Multi-dimensional vibration suppression control system and method for FLNG-LNGC platform with active and passive cooperative driving

Abstract

This invention relates to the field of automated control technology for marine engineering equipment, and discloses a multi-dimensional vibration suppression and adaptive loading and unloading control system and method for an active-passive coordinated drive FLNG / LNGC platform. The system includes sensing devices, a controller, and an execution loop. The sensing devices are used to collect the relative spatial pose signals of the two hulls and the spatial geometric configuration parameters of the transmission arm. The execution loop includes a proportional directional valve, a drive cylinder, and a proportional throttle valve connected in parallel between the drive cylinder and the accumulator group. The controller constructs a transient kinematic Jacobian matrix based on the collected data and calculates the pose stiffness transfer rate, thereby updating the cutoff frequency of the filter array. The pose signal is decomposed into low-frequency drift components and high-frequency wave excitation components. The controller generates active displacement tracking commands for the low-frequency components to control the proportional directional valve. This invention integrates low-frequency active displacement following and high-frequency passive unloading dissipation within the same hydraulic circuit, avoiding mechanical interference and reducing the risk of structural damage.

Description

Technical Field

[0001] This invention relates to the field of automated control technology for marine engineering equipment, specifically to a combined active and passive drive FLNG. Multidimensional vibration suppression control system and method for LNGC platform. Background Technology

[0002] In the development of offshore oil and gas resources, floating production storage and offloading (FPSO) units and transport vessels are usually moored side by side to transfer fluid media. In the marine environment, the two vessels will generate six degrees of freedom of spatial relative motion due to the combined effects of wind load, waves and ocean currents. This relative motion includes low-frequency spatial drift caused by ocean currents and tides, as well as high-frequency impacts and vibrations caused by short-period wind and waves. As a result, the transfer arm connecting the two vessels is subjected to complex alternating loads, which increases the risk of fatigue and damage to the mechanical structure.

[0003] Currently, control systems for transmission arms often struggle to reconcile low-frequency active displacement tracking with high-frequency passive load dissipation within the same execution loop. Conventional drive systems typically employ a single active tracking strategy. When faced with high-frequency wave excitation, the actuator still actively exerts force to counteract external impacts, resulting in pressure pulses within the hydraulic lines. The energy flows of active drive and external impacts are prone to interference in the physical loop. Furthermore, during the spatial movement of the transmission arm, when the mechanical link approaches the kinematic singular boundary, its structural stiffness in a specific direction decreases significantly. Most existing control logics do not establish a dynamic mapping relationship between control parameters and the transient physical stiffness of the transmission arm. If the system continues to execute active tracking commands when the mechanical structure is in a low-stiffness state, it can easily trigger singular divergence in the underlying calculation matrix, leading to control program shutdown. Moreover, it can easily exacerbate the physical deformation of the mechanical link in a fragile configuration, failing to meet the safety requirements for operations in complex sea conditions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for actively and passively co-driven FLNG. The multi-dimensional vibration suppression control system and method for LNGC platforms solves the problem that during the parallel operation of floating production storage and offloading (FPSO) units and transport vessels, the marine environment causes a complex relative motion between the two hulls, resulting in low-frequency spatial drift and high-frequency wave excitation. This causes alternating loads on the transmission arm, increasing the risk of fatigue and damage to the mechanical structure. Existing execution systems are difficult to reconcile low-frequency active displacement following and high-frequency passive unloading dissipation in the same loop. Furthermore, when the robotic arm approaches a kinematic singularity, causing a decrease in structural stiffness, the active drive execution is prone to causing control algorithm divergence or exacerbating mechanical damage.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A type of active-passive coordinated FLNG The LNGC platform multi-dimensional vibration suppression control system includes sensing devices, controllers, and execution loops; The sensing device includes a motion reference unit fixed to the floating production storage and unloading device and the transport ship, and an absolute encoder installed on the rotary joint of the transmission arm, which are used to collect the pose signal and spatial geometric configuration parameters of the six degrees of freedom in relative space, respectively. The execution circuit includes a proportional directional valve and a drive cylinder connected in sequence, and a proportional throttle valve arranged in parallel between the high-pressure working chamber of the drive cylinder and the accumulator group. The controller is communicatively connected to the sensing device, the proportional directional valve, and the proportional throttle valve, respectively.

[0006] Furthermore, the execution circuit also includes a variable pump and a pressure compensation valve; the output port of the variable pump is connected to the inlet of the pressure compensation valve, and the outlet of the pressure compensation valve is connected to the inlet of the proportional directional valve; the pressure compensation valve is provided with a pilot control feedback oil circuit, which is connected to the outlet pressure of the variable pump and the load pressure downstream of the proportional directional valve.

[0007] The pressure compensation valve dynamically adjusts the throttling opening based on the pressure difference between the control chambers on both sides of the internal valve core to compensate for fluctuations in load pressure and maintain the fluid pressure difference before and after the proportional directional valve as constant. Furthermore, the execution circuit is equipped with an extreme pressure safety relief valve connected in parallel at the inlet of the accumulator group; when the fluid pressure entering the accumulator group exceeds the preset physical safety boundary, the extreme pressure safety relief valve is passively opened to drain the excess fluid back to the oil tank to prevent the accumulator group from reaching the physical volume compression limit.

[0008] On the other hand, the present invention provides a method for active-passive coordinated driving of FLNG. The multi-dimensional vibration suppression control method for LNGC platforms includes the following steps: The relative spatial six-degree-of-freedom pose signal between the floating production storage and unloading device and the transport ship is acquired through the motion reference unit, and the spatial geometric configuration parameters of the transmission arm rotary joint are acquired through the absolute encoder. The controller constructs a transient kinematic Jacobian matrix based on the relative spatial six-degree-of-freedom pose signal and spatial geometric configuration parameters, and outputs the pose stiffness transfer rate. The controller has a preset Kalman bandpass filter array. The cutoff frequency of the Kalman bandpass filter array is updated according to the pose stiffness transfer rate. The Kalman bandpass filter array with the updated cutoff frequency is used to orthogonally decompose the pose signal into low-frequency drift components and high-frequency wave excitation components. An active displacement tracking command is generated for the low-frequency drift component and sent to the proportional directional valve to drive the drive cylinder to perform active displacement compensation. An impedance dissipation command is generated for the high-frequency wave excitation component and sent to the proportional throttle valve to adjust the damping characteristics between the high-pressure working chamber of the drive cylinder and the accumulator group, thereby achieving high-frequency vibration dissipation.

[0009] Furthermore, the process of acquiring the relative spatial six-degree-of-freedom pose signal between the floating production storage and offloading device and the transport vessel includes: The underlying acceleration signal is acquired through the motion reference unit; The controller performs a second integral operation on the underlying acceleration signal according to the set discrete-time scanning period, and uses a preset low-frequency blocking filter combined with a baseline correction algorithm to eliminate the displacement divergence error caused by integral accumulation; the controller calculates the fusion spatial installation offset parameters through the coordinate system homogeneous transformation matrix, and outputs the relative spatial six-degree-of-freedom pose signal at the connecting flange of the two ships.

[0010] Furthermore, the output pose stiffness transfer rate includes: The constructed transient kinematic Jacobian matrix is ​​subjected to singular value decomposition, and the maximum and minimum singular values ​​of the diagonal matrix in the orthogonal decomposition result are extracted; the ratio of the maximum singular value to the minimum singular value is calculated to obtain the pose stiffness transfer rate; and when the minimum singular value is lower than a preset regularization threshold, a preset minimum positive real number is assigned to the denominator to perform truncation protection to prevent division by zero overflow.

[0011] Further, updating the cutoff frequency of the Kalman bandpass filter array based on the pose stiffness transfer rate includes: The pose stiffness transfer rate is substituted into the natural exponential boundary function to perform a nonlinear mapping operation, and the dynamic cutoff frequency is calculated. The cutoff frequency is then restricted to a safe cutoff frequency lower limit greater than zero. Using this cutoff frequency as a boundary control parameter, the complementary low-pass and high-pass system transfer functions are reconstructed and transformed into discrete state-space equations to update the cutoff frequency of the Kalman bandpass filter array.

[0012] Furthermore, generating active displacement tracking commands includes: By transforming spatial coordinates and performing a pseudo-inverse operation of the Jacobian matrix in transient kinematics, the low-frequency drift components are converted into target angular displacements of each independent rotary joint and mapped to the target extension length of the drive cylinder; the actual displacement feedback value of the drive cylinder is obtained, and the position deviation is calculated based on the target extension length of the drive cylinder and the actual displacement feedback value of the drive cylinder; time difference operation is performed on the target extension length to obtain the target motion speed. The target movement speed and the position deviation are combined to generate a target flow control signal, which serves as the active displacement tracking command. The proportional directional valve adjusts its internal flow cross-sectional area according to the active displacement tracking command, thereby decoupling the low-frequency active drive flow entering the drive cylinder from external load fluctuations.

[0013] Furthermore, the instructions for generating impedance dissipation include: Time difference operation is performed on the high-frequency wave excitation component to obtain the high-frequency excitation velocity vector; the pose stiffness transfer rate is mapped to a nonlinear feedforward gain by combining the reference impedance gain and stiffness sensitivity factor with a nonlinear amplification factor; based on the dead zone compensation reference value to overcome the mechanical spring preload, the nonlinear feedforward gain and the L2 norm of the high-frequency excitation velocity vector are combined for calculation, and the impedance dissipation command is generated after performing amplitude saturation truncation processing.

[0014] Furthermore, when external high-frequency wave excitation induces fluid pressure pulses in the high-pressure working chamber, the proportional throttle valve dynamically expands its opening degree according to the impedance dissipation command; the high-frequency pressure pulse guides the hydraulic oil to flow through the proportional throttle valve to generate fluid shearing action and discharge into the accumulator group. By compressing the pre-charged gas inside the accumulator group, the mechanical impact kinetic energy is converted into gas compression potential energy for dissipation.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention constructs a transient kinematic Jacobian matrix and calculates the pose stiffness transfer rate to dynamically update the cutoff frequency of the Kalman bandpass filter array. This technical feature orthogonally decomposes complex external excitation signals into low-frequency drift components and high-frequency wave excitation components, establishing an adaptive correlation between frequency domain partitioning and the physical stiffness of the transmission arm. When the transmission arm approaches a kinematic singular position, causing a decrease in structural stiffness, the system reduces the cutoff frequency to classify more disturbed signals into the high-frequency excitation components, avoiding active countermeasure control on mechanical structures in a low-stiffness state, thereby preventing algorithm divergence and reducing the risk of mechanical damage.

[0016] 2. The execution circuit of this invention adopts a pilot control feedback oil circuit consisting of a variable pump and a pressure compensation valve. The pressure compensation valve is connected to the outlet pressure of the variable pump and the downstream load pressure of the proportional directional valve. The throttling opening is dynamically adjusted according to the pressure difference on both sides. This physical structure maintains a constant fluid pressure difference before and after the proportional directional valve, so that the low-frequency active drive flow entering the drive cylinder is only regulated by the control command. This constant difference pressure reduction mechanism physically isolates the load pressure fluctuation caused by external high-frequency waves, ensuring the absolute accuracy of the low-frequency spatial displacement following of the transmission arm.

[0017] 3. This invention sets up a proportional throttle valve in parallel between the high-pressure working chamber of the drive cylinder and the accumulator group, and generates an impedance dissipation command based on the high-frequency excitation velocity vector and the position stiffness transfer rate. When the external high-frequency mechanical impact is converted into a fluid pressure pulse in the cylinder chamber, the proportional throttle valve dynamically expands its opening degree, guiding the high-pressure fluid into the accumulator group, converting the mechanical impact kinetic energy into the compressive potential energy of the gas. This feature provides a release channel for high-frequency pulses in the same hydraulic system, eliminates the mechanical interference between low-frequency active drive and high-frequency passive unloading, and improves the fatigue resistance of the system to complex alternating loads. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall architecture of the active-passive coordinated drive multi-dimensional vibration suppression control system in an embodiment of the present invention; Figure 2 This is a flowchart of the real-time acquisition of multi-dimensional state parameters and the construction of spatial pose matrix in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the kinematic domain calculation and pose stiffness transfer rate construction in an embodiment of the present invention. Figure 4 This is a flowchart of signal domain and frequency domain manifold decoupling in an embodiment of the present invention; Figure 5 This is a flowchart illustrating the generation of active and passive energy flow commands in an embodiment of the present invention. Figure 6 This is a flowchart illustrating the non-interference energy allocation execution process in the fluid dynamics domain, as described in this embodiment of the invention. Figure 7 This is a time-displacement response curve diagram from an embodiment of the present invention; Figure 8 This is a transient pressure curve of the hydraulic cylinder working chamber in an embodiment of the present invention; Figure 9 This is a graph showing the combined response curve of adaptive cutoff frequency and stiffness transferability in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions in 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] Example 1 See attached document Figure 1 This invention provides a combined active and passive drive for FLNG. The LNGC platform multi-dimensional vibration suppression control system includes sensing devices, controllers, and execution loops.

[0021] The sensing equipment includes motion reference units and absolute encoders. The motion reference units are rigidly fixed to the decks of the floating production storage and offloading (FPSO) unit and the transport ship, respectively. The motion reference units are used to acquire the relative spatial six-degree-of-freedom pose signals between the two hulls. The absolute encoders are installed at each independent rotary joint of the transfer arm and are used to record the spatial geometric configuration parameters of the transfer arm.

[0022] The controller establishes communication connections with both the motion reference unit and the absolute encoder. Internally, the controller integrates a matrix operation coprocessor and digital signal processing circuitry to receive data from the underlying sensors and perform Jacobian matrix calculations and Kalman filter array derivation operations.

[0023] The execution circuit includes a variable displacement pump, a pressure compensation valve, a proportional directional valve, and a drive cylinder. The output port of the variable displacement pump is connected to the inlet of the pressure compensation valve via a pipeline. The outlet of the pressure compensation valve is connected to the inlet of the proportional directional valve. The first and second working ports of the proportional directional valve are respectively connected to the rodless and rod-side chambers of the drive cylinder. The mechanical output end of the drive cylinder is mechanically connected to the main drive node of the transmission arm.

[0024] The pressure compensation valve is equipped with a pilot control feedback oil circuit. The control chambers on both sides of the valve core inside the pressure compensation valve are respectively connected to the outlet pressure of the variable pump and the load pressure downstream of the proportional directional valve. This fluid topology maintains a constant fluid pressure difference across the directional valve in the physical circuit.

[0025] The actuation circuit also includes a proportional throttle valve and an accumulator group. The proportional throttle valve is connected in parallel between the high-pressure working chamber of the drive cylinder that bears the load and the accumulator group. The electrical control terminal of the proportional throttle valve is communicatively connected to the controller, and its fluid flow cross-sectional area changes according to the control commands issued by the controller.

[0026] During the operation of the overall control logic, the controller synchronously receives the relative motion pose vector from the motion reference unit and the joint angle position vector from the absolute encoder according to the set discrete-time scan cycle. The controller establishes the current input state matrix of the system based on the received multi-dimensional feature data.

[0027] The controller uses a forward kinematics model to analyze the joint angle position vectors, constructs a transient Jacobian matrix, and performs singular value decomposition on it. Based on the singular value decomposition results, the controller outputs the pose stiffness transfer rate, which characterizes the current load-bearing capacity of the robotic arm structure.

[0028] The controller substitutes the pose stiffness transferability into the natural exponential boundary function to calculate the safe cutoff frequency. Based on this safe cutoff frequency, the controller updates the parameters of the internal Kalman bandpass filter array. This filter array orthogonally decomposes the relative motion pose vector into low-frequency drift components and high-frequency wave excitation components.

[0029] For the low-frequency drift component obtained from the decomposition, the controller generates an active displacement tracking command and sends it to the electrical control terminal of the proportional directional valve. For the high-frequency wave excitation component, the controller generates an impedance dissipation command based on the attitude stiffness transfer rate and sends it to the proportional throttle valve.

[0030] The proportional directional valve receives the active displacement tracking command and adjusts the displacement opening of the internal valve core in real time. In conjunction with the pressure output of the variable pump, the pressure compensation valve isolates the pressure fluctuations of the downstream load, so that the flow rate entering the drive cylinder maintains a linear mapping relationship with the active displacement tracking command. The drive cylinder generates extension and retraction actions according to the controlled flow rate, driving the transmission arm to perform low-frequency displacement following.

[0031] When the high-frequency vibration of the external environment is transmitted to the drive cylinder along the transmission arm and causes fluid pressure pulses, the proportional throttle valve adjusts its opening degree according to the impedance dissipation command. The high-frequency pressure pulse guides the hydraulic oil through the proportional throttle valve into the accumulator group. The accumulator group consumes the pulse energy through the internal gas compression process, thus completing the high-frequency passive vibration suppression and unloading operation of the system.

[0032] Example 2 See attached document Figure 2 This embodiment provides a method for actively and passively co-driven FLNG. The multi-dimensional vibration suppression control method for LNGC platforms includes the following steps: Step 1: Acquire the relative spatial six-degree-of-freedom pose signal between the floating production storage and unloading device and the transport ship through the motion reference unit, and acquire the spatial geometric configuration parameters of the transmission arm rotary joint through the absolute encoder. Step 2: The controller constructs a transient kinematic Jacobian matrix based on the relative spatial six-degree-of-freedom pose signal and spatial geometric configuration parameters, and outputs the pose stiffness transfer rate. Step 3: The controller has a preset Kalman bandpass filter array. The cutoff frequency of the Kalman bandpass filter array is updated according to the pose stiffness transfer rate. The Kalman bandpass filter array with the updated cutoff frequency is used to orthogonally decompose the pose signal into low-frequency drift components and high-frequency wave excitation components. Step 4: Generate an active displacement tracking command for the low-frequency drift component and send it to the proportional directional valve to drive the drive cylinder to perform active displacement compensation. Step 5: Generate an impedance dissipation command for the high-frequency wave excitation component and send it to the proportional throttle valve to adjust the damping characteristics between the high-pressure working chamber of the drive cylinder and the accumulator group, thereby achieving high-frequency vibration dissipation.

[0033] In step 1 of this embodiment, the controller executes the state acquisition process according to the set discrete time scanning period. The specific value of the scanning period is usually determined according to the high frequency cutoff frequency of the environmental waves and Shannon sampling theorem. It can be set between 10 milliseconds and 50 milliseconds to balance the computing load of the controller and the dynamic response accuracy of the system. The state acquisition process mainly acquires the external relative motion pose data and the internal transient configuration parameters of the robotic arm.

[0034] Motion reference units installed on both sides of the floating production storage and offloading (FPSO) unit and transport vessel interact with the controller via an industrial Ethernet bus. In actual parallel operation environments, waves, wind loads, and ocean currents cause the two hulls to generate six degrees of freedom of independent spatial motion. This complex relative motion is the core excitation source causing mechanical fatigue in the transmission arm. Therefore, the controller reads the absolute roll, pitch, and bow angles, as well as heave, sway, and pitch accelerations of both hulls in each sampling cycle. To prevent dead zones caused by communication delays or packet loss, the controller performs timing alignment and validity checks after reading the data. If missing data is detected, the sampling value from the previous cycle is maintained to ensure the continuity of subsequent algorithm calls.

[0035] The controller's internal digital signal processing circuit performs a second integration operation on the acquired acceleration signal to obtain the corresponding translational displacement. To address the zero-point drift phenomenon during the acceleration signal integration process, the system employs a preset low-frequency blocking filter combined with a baseline correction algorithm to dynamically eliminate displacement divergence errors caused by long-term integration accumulation, ensuring the absolute accuracy of the spatial displacement calculation.

[0036] After acquiring the absolute pose data of both hulls after processing, the system needs to establish a unified reference benchmark. The controller typically establishes a local reference coordinate system with the center of the connecting flange of the floating production storage and offloading (FPSO) unit as the origin. The relative spatial pose vector at the connecting flange of the two vessels is calculated using a homogeneous transformation matrix. This homogeneous transformation matrix incorporates the spatial installation offset parameters from the centers of the two vessels to their respective flanges. The relative spatial pose vector is defined as follows: Its mathematical expression is: ; In the formula, Represents the current discrete sampling time; It represents the relative translational displacement along the longitudinal X-axis of the reference coordinate system (usually defined as the bow and stern direction of the hull); This indicates the relative translational displacement along the transverse Y-axis (port / ship direction); It represents the relative translational displacement along the vertical Z-axis (opposite to gravity); Indicates the relative roll angle about the X-axis; Indicates the relative pitch angle about the Y-axis; Indicates the relative bow roll angle about the Z-axis; superscript This represents the matrix transpose operation.

[0037] This relative spatial pose vector fully characterizes the spatial displacement constraint state at the connecting flange from a physical perspective and serves as input data for the system to identify the excitation intensity of the external environment.

[0038] In addition to external excitation input, the posture of the transmission arm itself directly determines its mechanical characteristics for bearing and transmitting loads. Absolute encoders distributed at each rotating joint of the transmission arm collect the current absolute angular displacement of each joint under the synchronous trigger signal of the controller. The control system uses absolute encoders instead of incremental encoders, mainly because offshore platforms are prone to power outages or abnormal shutdowns. The absolute feedback mechanism can ensure that the system does not need to perform the action of finding the physical zero point when resetting and restarting, thereby eliminating control delay and potential collision risks.

[0039] The controller reads the absolute angular displacement data of all independent moving joints and constructs joint angular position vectors according to the mechanical topology order from the transmission arm base to the end connecting flange. The joint angular position vector is defined as follows: Its mathematical expression is: ; In the formula, This refers to the total number of independent active degrees of freedom contained in the transmission arm. The specific value is determined by the mechanical structure of the transmission arm used, and is usually 6 or less. to Representing the first joint to the second joint respectively Joints at discrete sampling times The actual absolute angular displacement data.

[0040] The joint angle position vector is used in the control logic to characterize the real-time spatial geometric attitude of the transmission arm, and serves as the data basis for subsequent solving of the transient forward kinematics model and determining the physical stiffness boundary of the structure.

[0041] After completing the construction of the two sets of core data, the controller stores the relative spatial pose vector and joint angle position vector into the internal register for the calculation module to call in the next control cycle.

[0042] See attached document Figure 3 In step 2 of this embodiment, the controller executes a kinematic domain calculation program based on the joint angle position vector obtained in the preceding process. This calculation program is used to evaluate the structural resistance of the transient spatial configuration of the transmission arm to external excitation loads in real time.

[0043] The controller extracts the joint angle position vector at the current discrete sampling time. The mapping relationship between the spatial velocities of the bottom joints of the robotic arm and the Cartesian spatial velocities of the end effector is established by the kinematic Jacobian matrix. The controller has a pre-set kinematic model representing the physical dimensions and constraint relationships of the transmission arm. Based on this model, the controller calculates the corresponding kinematic Jacobian matrix in real time. From the perspective of mechanical principles, the Jacobian matrix not only determines the proportional relationship of motion transmission, but also, according to the principle of virtual work, its transpose constitutes a static mechanical mapping network for the transmission of external environmental forces to internal joint torques. Therefore, analyzing this matrix is ​​fundamental to determining the system's resistance to disturbances. Its mathematical expression is: ; In the formula, This is the transient kinematic Jacobian matrix corresponding to the current joint angle position vector; to This represents the partial derivatives of the motion components of the connecting flange at the end of the transmission arm in the six directions of Cartesian space with respect to the angular displacements of each independent rotary joint; The variable representing the translational degrees of freedom of the end flange in the spatial reference coordinate system; The variable representing the spatial rotational degrees of freedom of the end flange; to Represents the individual joint variables of the transmission arm from the base to the end; This represents the total number of independent active degrees of freedom of the transmission arm.

[0044] In the specific computational stage, the controller iterates the homogeneous transformation matrix between adjacent links and calculates the partial derivatives of each joint degree of freedom variable to complete the dynamic filling and updating of the high-dimensional Jacobian matrix.

[0045] In this embodiment, to quantify the isotropic degree of structural stiffness of the current spatial configuration of the transmission arm, the controller calculates the kinematic Jacobian matrix. Singular value decomposition is performed. The controller's underlying computational unit transforms the aforementioned high-dimensional Jacobian matrix into a product of an orthogonal matrix and a diagonal matrix using an iterative algorithm. The controller extracts non-negative diagonal elements from the diagonal matrix and selects the largest and smallest singular values. Based on the selected singular value parameters, the controller calculates the pose stiffness transfer rate for the current pose. The pose stiffness transmissibility is mathematically equivalent to the condition number of the Jacobian matrix, and its mathematical expression is: ; In the formula, For the current discrete sampling time The pose stiffness transfer rate; For the kinematic Jacobian matrix The maximum singular value; This is the corresponding minimum singular value.

[0046] In practical engineering calculations, to prevent the controller from crashing due to division by zero overflow caused by the minimum singular value being equal to zero when the transmission arm is in an absolutely singular position, the controller internally incorporates regularization truncation protection for the calculation program. Specifically, the system presets a very small positive real number as the regularization threshold, and the value range of this regularization threshold is typically set to 10. -6 Up to 10 -4 The specific value depends on the floating-point precision of the controller and the physical dimensions of the transmission arm. When detected... When the value is below the regularization threshold, the regularization threshold is directly assigned to the denominator for calculation, thereby eliminating the risk of dead zones that may exist during the algorithm's runtime.

[0047] Based on the physical characteristic mapping relationship, the controller calculates... The numerical value directly characterizes the transmission arm's ability to resist external mechanical deformation. Specifically, when When the value approaches 1, it indicates that the mechanical transmission of the current spatial configuration of the transmission arm tends to be uniform in all directions, and the overall torsional and bending stiffness of the structure is at a high level. When the numerical value increases abnormally, it indicates that the mechanical connector of the transmission arm is approaching the kinematic singular boundary of full extension or extreme folding. The mechanical arm near the singular boundary loses some degrees of freedom of movement, and its structural stiffness in a specific direction decreases sharply, making it unable to effectively resist the high-frequency excitation force caused by the waves.

[0048] After the controller completes the above calculations, it stores the pose stiffness transfer rate into the global register. This parameter transcends the realm of pure kinematics and serves as a core control variable that constrains the frequency boundary of the subsequent signal filtering module, participating in the coordinated allocation of the system's active and passive energy flows.

[0049] In step 3 of this embodiment, refer to the appendix. Figure 4 To address the mechanical interference caused by active drive and passive unloading actions within the same hydraulic actuation circuit, the controller employs a frequency domain isolation strategy to process the input external environmental excitation signal. This process relies on the pose stiffness transferability calculated by the preceding module.

[0050] In traditional linear filtering mapping, when the robotic arm approaches a kinematic singularity, causing a sharp increase in pose stiffness transfer rate, the calculated filter cutoff frequency is prone to becoming negative or zero, leading to divergence or shutdown of the underlying control algorithm. To avoid this risk of computational singularity, a dynamic cutoff frequency mapping equation based on the natural exponential decay characteristic is constructed within the controller.

[0051] In practice, the controller extracts the pose stiffness transfer rate at the current discrete sampling moment. Then, substitute it into the mapping equation to perform the operation. The actual calculation formula is: ; In the formula, The dynamic cutoff frequency calculated at the current moment; The preset cutoff frequency for the system. This preset cutoff frequency is determined based on the dominant frequency of the conventional wave spectrum in the operating sea area, and its value is usually set between 0.5 Hz and 1.5 Hz; The lower limit of the safety cutoff frequency is set to ensure the basic operating conditions of the filter array. This lower limit value must be a positive real number that is strictly greater than zero, and is usually taken as 0.1 Hz to 0.3 Hz. This is a stiffness-sensitive penalty coefficient used to adjust the rate of frequency decay. It can be set within the range of 0.5 to 2.0 according to the actual system response requirements. is the base of the natural logarithm.

[0052] From the mathematical convergence characteristics of the above formula, it can be seen that regardless of the pose stiffness transfer rate... How does the exponential term surge near the singularity? All of them exhibit an asymptotic characteristic that approaches zero infinitely. This nonlinear asymptotic saturation function ensures the dynamic cutoff frequency. Under any extreme operating conditions, it is limited to the lower limit of the safe cutoff frequency. This avoids algorithm crashes caused by abnormal truncation frequency and ensures the continuous stability of the controller operation.

[0053] Obtain a safe and reliable dynamic truncation frequency Subsequently, the controller uses this as a boundary control parameter to update the state-space model matrix of the internal adaptive Kalman bandpass filter array in real time. The controller then uses the acquired relative spatial pose vector... The input is fed into the updated filter array. At the specific algorithm construction level, the controller uses... As a core variable, the transfer functions of complementary low-pass and high-pass systems are reconstructed in real time, and these transfer functions are transformed into discrete state-space equations embedded in the prediction and update mechanism of the Kalman filter. The filter array thereby orthogonally decomposes the original input broadband relative motion pose data into two independent frequency band signal manifolds. The specific mathematical decoupling relationship is expressed as follows: ; In the formula, The low-frequency drift component is extracted after low-pass filtering, which mainly includes long-period slow drift displacement data of the hull caused by ocean currents and tides. The high-frequency wave excitation component is extracted after high-pass filtering, and it mainly includes data on high-frequency turbulence and impact vibration of the ship caused by short-period wind and waves.

[0054] This frequency domain decoupling process establishes an adaptive correlation between signal frequency domain partitioning and physical stiffness in the control logic. When the transmission arm tends towards a kinematically singular position and the structural stiffness decreases, the increase in pose stiffness transferability leads to a dynamic truncation frequency. Consequently, it decreases. This parameter adjustment causes more pose signals in the middle and edge frequency bands to be classified as high-frequency wave excitation components. In this dynamic frequency domain delimitation mechanism, the controller pre-directs more external excitation energy to the subsequent high-frequency passive impedance dissipation circuit. This significantly reduces the displacement tracking burden on the active drive circuit and prevents mechanical damage to the transmission arm, which is in a low-stiffness and vulnerable state, due to forced active tracking.

[0055] In the state-space prediction and correction stage, the underlying solution unit dynamically minimizes the covariance of the pose state estimation error through continuous iterative multiplication and addition operations of the system state transition matrix and the observation matrix, thereby completing the orthogonal decomposition and outputting the two separated frequency domain components in parallel to the instruction generation module.

[0056] See attached document Figure 5 In this embodiment, the controller calculates the active control commands and passive unloading commands required by the fluid execution loop based on the low-frequency drift component and high-frequency wave excitation component separated from the preceding process. This command generation process realizes the data mapping from the signal processing domain to the fluid physical domain.

[0057] In step 4 of this embodiment, for the separated low-frequency drift component, the system needs to drive the transmission arm to perform slow position following to compensate for the basic relative displacement between the two hulls. During this process, the controller extracts the low-frequency drift component. The controller then substitutes this into a pre-set inverse kinematics model of the transmission arm. Through spatial coordinate transformation and pseudo-inverse operation of the Jacobian matrix, the controller converts the low-frequency drift component in Cartesian space into the target angular displacement of each independent rotary joint. Based on the geometric constraints of the underlying mechanical linkage, the controller further maps the above target angular displacement to the target extension length of the drive cylinder. To form a tight closed-loop control link, the system typically incorporates a linear displacement sensor to synchronously acquire the current actual displacement feedback value of the drive cylinder and calculate the positional deviation between the actual displacement and the target extension length. Using this positional deviation, the controller generates an active displacement tracking command. This command is physically mapped to a target flow control signal that controls the opening degree of the proportional directional valve. The calculation formula is as follows: ; In the formula, The target flow control signal is calculated at the current discrete sampling time. The effective working area of ​​the hydraulic cylinder's working chamber; The first derivative of the target's stretching length with respect to time is the target's velocity. The position loop proportional control gain is typically set between 5.0 and 15.0 based on the system's full load mass and the hydraulic natural frequency. This refers to the real-time deviation between the target extension / retraction length of the hydraulic cylinder and the actual displacement feedback value.

[0058] The combination of the feedforward term and the deviation adjustment term ensures linear tracking of the flow output and the low-frequency slow-drift displacement. This closed-loop control law, by continuously updating the correction quantity, can quickly decay the position tracking deviation and effectively suppress motion overshoot.

[0059] After completing the low-frequency displacement tracking calculation, properly handling the high-frequency excitation energy becomes the key to the system's collision avoidance and fatigue resistance.

[0060] In step 5 of this embodiment, when dealing with the high-frequency wave excitation component, the system no longer actively counteracts the force, but instead absorbs and dissipates external impact energy by adjusting the acoustic impedance of the fluid circuit. From the perspective of fluid dynamics, by changing the throttling area in the hydraulic circuit, the compliance of the actuator with external forces can be dynamically adjusted, making it exhibit physical characteristics similar to a variable spring damper. To match the passive compliance of the actuator circuit with the transient structural stiffness of the robotic arm, the controller introduces a nonlinear feedforward gain mechanism.

[0061] The controller controls the high-frequency wave excitation components. Perform time difference calculation to obtain the corresponding high-frequency excitation velocity vector. The pose stiffness transfer rate is obtained by combining the calculations from the preceding modules. The controller calculates the nonlinear feedforward gain. The mapping equation for this gain is: ; In the formula, This is a real-time nonlinear feedforward gain; This is the reference impedance gain, typically ranging from 0.1 to 0.5. This is a nonlinear amplification factor used to define the gain penalty when stiffness deteriorates, and is usually set between 1.0 and 3.0; As a stiffness sensitivity factor, its value is usually set between 0.5 and 2.0, and the specific value is determined by the dynamic response bandwidth of the proportional throttle valve. is the base of the natural logarithm.

[0062] When the transmission arm is in the optimal compressive configuration, i.e. When the value approaches 1, the feedforward gain remains at the baseline level; when the mechanical structure approaches a singularity, it causes... As the gain increases, the feedforward gain rises exponentially. Based on the aforementioned nonlinear feedforward gain, the controller calculates the high-frequency passive impedance dissipation command sent to the proportional throttle valve. This instruction determines the transient fluid flow cross-sectional area of ​​the proportional throttle valve, and its calculation formula is as follows: ; In the formula, This is the control voltage or current command output to the proportional throttle valve; This is a dead zone compensation reference value used to overcome the preload of the valve core mechanical spring. This reference value is usually obtained by measuring the pressure-flow opening and overflow characteristic curve of the proportional throttle valve during the offline commissioning phase of the system. The L2 norm of the high-frequency excitation velocity vector represents the overall intensity of the current external excitation. To prevent excessively large calculated command amplitudes under extreme sea conditions from burning out the valve solenoid coil or causing controller output overflow, the controller performs a check before outputting to the hardware port. Perform amplitude saturation cutoff to strictly limit it to the rated input range of the proportional throttle valve (e.g., 0 to 10V or 4 to 20mA).

[0063] Through this instruction generation logic, the system achieves advance prediction of physical states. When the external high-frequency excitation speed increases and the resistance of the transmission arm structure decreases, the controller increases the impedance dissipation command to expand the opening of the proportional throttle valve in advance. This operation of reducing local flow resistance causes the high-pressure fluid in the cylinder cavity caused by external high-frequency excitation to be preferentially discharged into the accumulator group through the proportional throttle valve. The nitrogen gas inside the accumulator group is compressed and contracts, converting the mechanical impact kinetic energy into gas compression potential energy for absorption, effectively avoiding fatigue damage to hydraulic pipelines and mechanical components caused by high-frequency alternating loads. In this embodiment, the controller synchronously outputs the above-mentioned active displacement tracking command and impedance dissipation command within one scan cycle to drive the underlying fluid topology structure to perform actions.

[0064] See attached document Figure 6 In this embodiment, the control system converts the electrical signal commands generated at the front end into mechanical actions at the physical level through a specific fluid topology. This execution process needs to be compatible with both low-frequency active drive and high-frequency passive unloading within the same hydraulic circuit to avoid mechanical interference between the two energy flows.

[0065] In the complex sea conditions of parallel operations at sea, external waves exert alternating load forces on the transmission arm, causing nonlinear fluctuations in the fluid pressure inside the drive cylinder. According to basic fluid mechanics principles, if the pressure difference across the throttle valve is unstable, the actual output flow rate will deviate from the expected target, thus compromising the tracking accuracy of low-frequency displacement. To address this interference problem, the system incorporates a differential pressure reduction structure consisting of a pressure compensation valve and a directional valve in the execution circuit.

[0066] The proportional directional valve receives the active displacement tracking command issued by the controller and adjusts the internal flow cross-sectional area in real time. The pressure compensation valve is set between the variable pump and the proportional directional valve. The two control chambers on both sides of its internal valve core are respectively connected to the outlet pressure of the variable pump and the load pressure downstream of the proportional directional valve. Through this pilot control feedback oil circuit physical feedback mechanism, the valve core of the pressure compensation valve dynamically adjusts its throttling opening according to the pressure on both sides. This action compensates for the fluctuation of the load pressure in real time in the physical circuit, so that the fluid pressure difference before and after the proportional directional valve is maintained at a constant state.

[0067] Under the physical boundary condition of constant pressure difference, the actual flow rate entering the drive cylinder is calculated using the fluid orifice flow equation, and its mathematical expression is: ; In the formula, This represents the actual fluid flow rate into the drive cylinder at the current moment. This is the fluid flow coefficient of the directional valve; This represents the current fluid flow cross-sectional area of ​​the directional valve; The constant fluid pressure difference is determined by the stiffness of the internal mechanical spring of the pressure compensation valve. This pressure difference is usually set in the range of 0.5 MPa to 1.5 MPa to balance the resolution of flow regulation and the pressure loss of the system. This refers to the fluid density of the hydraulic oil.

[0068] As can be seen from the above fluid mechanics formulas, due to Forced to remain constant by hardware loops, actual flow Only the flow cross-sectional area of ​​the reversing valve It exhibits a linear proportional relationship and is completely decoupled from fluctuations in the external load. This forced decoupling mechanism ensures that the low-frequency active drive energy flow entering the drive cylinder is not disturbed by high-frequency wave excitation, guaranteeing the spatial displacement following accuracy of the transmission arm under complex loads.

[0069] After completing the hardware decoupling of the low-frequency drive flow, another core technical challenge the system faces is how to provide a safe and reliable release channel for the isolated high-frequency impact energy.

[0070] When high-frequency wave vibrations from the external environment are transmitted to the drive cylinder along the mechanical structure of the transmission arm, the mechanical impact force exceeding the normal load-bearing threshold forces the piston of the drive cylinder to exhibit a high-frequency vibration tendency. This tendency is rapidly converted into fluid pressure pulses within the closed hydraulic cylinder chamber. If this high-pressure fluid is not released in time, it may cause the mechanical connecting rod to yield and deform.

[0071] To dissipate this high-frequency impact energy, a proportional throttle valve, installed in parallel between the high-pressure working chamber of the drive cylinder and the accumulator group, adjusts its opening degree according to the impedance dissipation command. When a high-frequency pressure pulse occurs in the cylinder chamber, the high-pressure hydraulic oil is guided to flow through the proportional throttle valve. As the hydraulic oil flows through the narrow cross-section of the valve orifice, fluid shearing occurs, converting some of the mechanical energy into fluid heat energy for dissipation.

[0072] Hydraulic oil flowing through the proportional throttle valve enters the accumulator group. The accumulator group is pre-charged with high-pressure gas. The pre-charge pressure of the gas inside the accumulator is typically set between 60% and 80% of the system's average operating pressure, based on the static pressure equilibrium point under no-load and full-load conditions, to ensure the airbag has optimal volume contraction margin. The hydraulic oil entering the accumulator compresses the internal gas, converting the transient kinetic energy of high-frequency impacts into the gas's compressive potential energy. The physical model of this passive energy dissipation and absorption process can be expressed as: ; In the formula, For the time interval of being impacted by high-frequency waves The total energy absorbed and dissipated by the internal system; The transient fluid pressure at the inlet of the accumulator group; The transient fluid flow rate into the accumulator group via the proportional throttle valve.

[0073] To prevent the accumulator assembly from reaching its physical volume compression limit due to continuous energy absorption under extreme sea conditions, which could lead to hydraulic lock-up and a physical dead zone caused by a pressure surge, an extreme pressure safety relief valve is connected in parallel at the inlet of the accumulator assembly. When the system pressure exceeds the set physical safety boundary, this relief valve is passively opened, directly releasing excess high-pressure fluid back to the oil tank, thus establishing a hardware-level explosion-proof barrier.

[0074] Through this physical mechanism that combines throttling dissipation and elastic energy storage, the execution loop exhibits dynamically variable impedance characteristics based on real-time control commands. This allows the system to smoothly mitigate the destructive energy caused by high-frequency wave excitation while maintaining low-frequency displacement tracking. Simultaneously, an independent fluid circulation auxiliary system continuously provides oil filtration and heat exchange cooling services to the pipelines to maintain the stability of the physical and chemical properties of the hydraulic medium during long-term operation. The control system cyclically executes all the above calculations and execution processes throughout the entire discrete-time scan cycle, achieving continuous control of multi-dimensional vibration suppression and adaptive loading / unloading.

[0075] See attached document Figure 7 To be continued Figure 9 This document provides a specific application example for parallel export operations in deepwater oil and gas fields.

[0076] In this embodiment, the operating environment is set in a deep-water area of ​​the South China Sea, with a sea state of level 4 and typical wave characteristics of a significant wave height of 3.2 meters and a peak period of 8.5 seconds. The floating production storage and offloading (FPSO) unit and the shuttle tanker are moored side by side, and crude oil is transferred between them via a transfer arm with six independent rotating joints. The system's control reference cutoff frequency is set to 0.12 Hz, the constant differential pressure setting of the pressure compensation valve is set to 1.0 MPa, and the pre-charge nitrogen pressure of the accumulator group is set to 12 MPa (the system's rated operating pressure is 20 MPa).

[0077] Once parallel operation commences, the motion reference units on both sides capture the six-degree-of-freedom relative pose changes of the two hulls in real time under complex sea conditions. After receiving the sensor data, the controller uses its internal Jacobian matrix to assess the spatial pose stiffness of the current transmission arm. Under the impact of a long-period surge, the Kalman bandpass filter array inside the controller quickly decouples the relative displacement signal: the separated low-frequency drift component (frequency below 0.12 Hz, representing the slow relative pulling of the two hulls due to ocean currents) is converted into an active displacement tracking command. This command drives the proportional directional valve in the hydraulic circuit to open slowly, coordinating with the variable pump to output a stable flow rate. Under the forced constant pressure difference of the pressure compensation valve, the drive cylinder smoothly extends 1.5 meters, compensating for the basic displacement deviation between the two hulls, without interference from high-frequency waves throughout the process. Simultaneously, the separated high-frequency wave excitation component (frequency concentrated between 0.11 and 0.25 Hz, representing the instantaneous impact of waves on the hulls) is converted into an impedance dissipation command. When wave impact causes a surge in force on the transmission arm and the internal pressure of the drive cylinder instantaneously exceeds 18 MPa, the proportional throttle valve rapidly expands its internal flow cross-sectional area according to this command. High-pressure hydraulic fluid is smoothly discharged into the accumulator group, where the transient kinetic energy of the wave impact is converted into potential energy by compressing the internal nitrogen gas. As the wave crest recedes, the energy in the accumulator is then gently released through the throttle valve, thus flexibly mitigating the destructive force that could lead to the yielding of the mechanical connecting rod without changing the macroscopic position of the drive cylinder.

[0078] The simulation input source uses the JONSWAP standard wave spectrum to simulate the external excitation input of the aforementioned sea state level 4. The comparison objects are set as "System A: Traditional rigid PID position closed-loop control system" and "System B: Active-passive coordinated drive multi-dimensional vibration suppression control system proposed in this invention". The simulation period is set to 200 seconds, and the discrete-time scan period is set to 20 milliseconds.

[0079] The displacement tracking accuracy comparison and verification can be visually observed in the time-domain displacement tracking curves. System A, unable to isolate high-frequency excitation, exhibits severe sawtooth oscillations in the actual displacement curve of its driving cylinder near the target low-frequency trajectory, with a maximum position tracking error reaching ±245 mm. This high-frequency mechanical vibration easily leads to fatigue fracture of the transmission arm joint. Conversely, the actual displacement trajectory of System B closely matches the target low-frequency slow drift displacement, with a smooth trajectory and no obvious burrs. The maximum position tracking error is effectively converged to within ±18 mm. This demonstrates the extremely high anti-interference capability of the constant-difference pressure reduction circuit combined with the frequency domain decoupling algorithm in flow control.

[0080] The pressure pulse comparison verification of the core actuator is a key indicator for measuring the system's shock resistance. During the concentrated impact phase of the wave cluster from 45 to 55 seconds, the working chamber pressure of the drive cylinder in System A exhibited an extremely steep pulse peak, reaching a maximum of 31.5 MPa, far exceeding the yield limit of conventional pipeline designs. By applying the technical solution of this invention, the variable acoustic impedance network composed of the proportional throttle valve and accumulator group in System B played a crucial passive unloading role. The working chamber pressure curve of System B exhibited a smooth, obtuse-angle transition, with the maximum pressure strictly limited to a safe threshold of 19.2 MPa. The pressure peak was reduced by 39% compared to the traditional solution, completely avoiding the risk of hydraulic component lock-up and rupture.

[0081] Frequency domain decoupling and stiffness adaptive verification: The dual Y-axis curves clearly demonstrate the algorithm's self-protection mechanism. At the 110-second mark of the simulation, the transmission arm is forced to approach a kinematic singularity, and the pose stiffness transferability... The frequency rapidly climbs from a baseline of 2.1 to 15.4. The dynamic cutoff frequency of the controller is recorded under the mapping of the natural exponential boundary function. The frequency rapidly and smoothly decays from 0.12 Hz to the set safety lower limit of 0.05 Hz. This dynamic offset confirms that the system can actively expand the frequency band of high-frequency passive unloading when the mechanical structure is in a low-stiffness and vulnerable period, guiding more wave damage energy to the energy storage device for dissipation, thereby achieving an adaptive closed loop between signal processing and physical load-bearing capacity.

[0082] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A PCC driven FLNG A multi-dimensional vibration control system for an LNGC platform, characterized by, Includes sensing devices, controllers, and execution loops; The sensing device includes a motion reference unit fixed to the floating production storage and unloading device and the transport ship, and an absolute encoder installed on the rotary joint of the transmission arm, which are used to collect the pose signal and spatial geometric configuration parameters of the six degrees of freedom in relative space, respectively. The execution circuit includes a proportional directional valve and a drive cylinder connected in sequence, and a proportional throttle valve arranged in parallel between the high-pressure working chamber of the drive cylinder and the accumulator group. The controller is communicatively connected to the sensing device, the proportional directional valve, and the proportional throttle valve, respectively.

2. The PMSFLNG of claim 1 A multi-dimensional vibration control system for an LNGC platform, characterized by, The execution circuit also includes a variable pump and a pressure compensation valve; the output port of the variable pump is connected to the inlet of the pressure compensation valve, and the outlet of the pressure compensation valve is connected to the inlet of the proportional directional valve; the pressure compensation valve is provided with a pilot control feedback oil circuit, which is connected to the outlet pressure of the variable pump and the load pressure downstream of the proportional directional valve. The pressure compensation valve dynamically adjusts the throttling opening based on the pressure difference between the control chambers on both sides of the internal valve core to compensate for fluctuations in load pressure and maintain the fluid pressure difference before and after the proportional directional valve as constant.

3. The active-passive coordinated drive FLNG according to claim 1 The LNGC platform multi-dimensional vibration suppression control system is characterized by, The execution circuit is equipped with an extreme pressure safety relief valve connected in parallel at the inlet of the accumulator group; when the fluid pressure entering the accumulator group exceeds the preset physical safety boundary, the extreme pressure safety relief valve is passively opened to drain the excess fluid back to the oil tank to prevent the accumulator group from reaching the physical volume compression limit.

4. A method for active-passive coordinated driving of FLNG The multi-dimensional vibration suppression control method for the LNGC platform, applied to the active-passive coordinated drive FLNG as described in any one of claims 1-3. The LNGC platform multi-dimensional vibration suppression control system is characterized by, Includes the following steps: The relative spatial six-degree-of-freedom pose signal between the floating production storage and unloading device and the transport ship is acquired through the motion reference unit, and the spatial geometric configuration parameters of the transmission arm rotary joint are acquired through the absolute encoder. The controller constructs a transient kinematic Jacobian matrix based on the relative spatial six-degree-of-freedom pose signal and spatial geometric configuration parameters, and outputs the pose stiffness transfer rate. The controller has a preset Kalman bandpass filter array. The cutoff frequency of the Kalman bandpass filter array is updated according to the pose stiffness transfer rate. The Kalman bandpass filter array with the updated cutoff frequency is used to orthogonally decompose the pose signal into low-frequency drift components and high-frequency wave excitation components. An active displacement tracking command is generated for the low-frequency drift component and sent to the proportional directional valve to drive the drive cylinder to perform active displacement compensation. An impedance dissipation command is generated for the high-frequency wave excitation component and sent to the proportional throttle valve to adjust the damping characteristics between the high-pressure working chamber of the drive cylinder and the accumulator group, thereby achieving high-frequency vibration dissipation.

5. The active-passive coordinated drive FLNG according to claim 4 The multi-dimensional vibration suppression control method for LNGC platforms is characterized by, The process of acquiring the relative spatial six-degree-of-freedom pose signal between the floating production storage and offloading device and the transport vessel includes: The underlying acceleration signal is acquired through the motion reference unit; The controller performs a second integral operation on the underlying acceleration signal according to the set discrete-time scanning period, and uses a preset low-frequency blocking filter combined with a baseline correction algorithm to eliminate the displacement divergence error caused by integral accumulation; the controller calculates the fusion spatial installation offset parameters through the coordinate system homogeneous transformation matrix, and outputs the relative spatial six-degree-of-freedom pose signal at the connecting flange of the two ships.

6. The active-passive coordinated drive FLNG according to claim 4 The multi-dimensional vibration suppression control method for LNGC platforms is characterized by, The output pose stiffness transfer rate includes: The constructed transient kinematic Jacobian matrix is ​​subjected to singular value decomposition, and the maximum and minimum singular values ​​of the diagonal matrix in the orthogonal decomposition result are extracted; the ratio of the maximum singular value to the minimum singular value is calculated to obtain the pose stiffness transfer rate; and when the minimum singular value is lower than a preset regularization threshold, a preset minimum positive real number is assigned to the denominator to perform truncation protection to prevent division by zero overflow.

7. The active-passive coordinated drive FLNG according to claim 4 The multi-dimensional vibration suppression control method for LNGC platforms is characterized by, Updating the cutoff frequency of the Kalman bandpass filter array based on the pose stiffness transfer rate includes: The pose stiffness transfer rate is substituted into the natural exponential boundary function to perform a nonlinear mapping operation, and the dynamic cutoff frequency is calculated. The cutoff frequency is then restricted to a safe cutoff frequency lower limit greater than zero. Using this cutoff frequency as a boundary control parameter, the complementary low-pass and high-pass system transfer functions are reconstructed and transformed into discrete state-space equations to update the cutoff frequency of the Kalman bandpass filter array.

8. The active-passive coordinated drive FLNG according to claim 4 The multi-dimensional vibration suppression control method for LNGC platforms is characterized by, The generation of active displacement tracking commands includes: By transforming spatial coordinates and performing a pseudo-inverse operation of the Jacobian matrix in transient kinematics, the low-frequency drift components are converted into target angular displacements of each independent rotary joint and mapped to the target extension length of the drive cylinder; the actual displacement feedback value of the drive cylinder is obtained, and the position deviation is calculated based on the target extension length of the drive cylinder and the actual displacement feedback value of the drive cylinder; time difference operation is performed on the target extension length to obtain the target motion speed. The target movement speed and the position deviation are combined to generate a target flow control signal, which serves as the active displacement tracking command. The proportional directional valve adjusts its internal flow cross-sectional area according to the active displacement tracking command, thereby decoupling the low-frequency active drive flow entering the drive cylinder from external load fluctuations.

9. The active-passive coordinated drive FLNG according to claim 4 The multi-dimensional vibration suppression control method for LNGC platforms is characterized by, The commands for generating impedance dissipation include: Time difference operation is performed on the high-frequency wave excitation component to obtain the high-frequency excitation velocity vector; the pose stiffness transfer rate is mapped to a nonlinear feedforward gain by combining the reference impedance gain and stiffness sensitivity factor with a nonlinear amplification factor; based on the dead zone compensation reference value to overcome the mechanical spring preload, the nonlinear feedforward gain and the L2 norm of the high-frequency excitation velocity vector are combined for calculation, and the impedance dissipation command is generated after performing amplitude saturation truncation processing.

10. The active-passive coordinated drive FLNG according to claim 9 The LNGC platform multi-dimensional vibration suppression control system is characterized by, When external high-frequency wave excitation induces fluid pressure pulses in the high-pressure working chamber, the proportional throttle valve dynamically expands its opening degree according to the impedance dissipation command; the high-frequency pressure pulse guides the hydraulic oil to flow through the proportional throttle valve to generate fluid shearing action and discharge into the accumulator group. By compressing the pre-charged gas inside the accumulator group, the mechanical impact kinetic energy is converted into gas compression potential energy for dissipation.

Images