A control method and system for a five-degree-of-freedom marine crane in a non-inertial frame of reference.

Abstract

This invention provides a control method and system for a five-degree-of-freedom marine crane in a non-inertial frame, relating to the field of crane automatic control technology. The method includes: acquiring the real-time state variables of the crane in a non-inertial frame; inputting the state variables into an adaptive radial basis function neural network to obtain a lumped disturbance estimate, the neural network being designed based on a five-degree-of-freedom dynamic model established in the ship's coordinate system using the Lagrange equations and the principle of virtual work, considering the ship's six-degree-of-freedom motion and installation position offset; calculating a tracking error function based on the state variables and the desired trajectory; inputting the disturbance estimate and the error function into a feedback controller to calculate and generate a control torque, which is then output to the actuator to drive the crane to track the desired trajectory and suppress load sway. This invention achieves direct modeling and control in a non-inertial frame, solving the coordinate transformation disturbance and singularity problems caused by inertial frame modeling, and improving trajectory tracking accuracy and disturbance rejection robustness.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology for cranes, and in particular to a control method and system for a five-degree-of-freedom marine crane in a non-inertial frame. Background Technology

[0002] Marine cranes are specialized lifting equipment used for tasks such as transferring cargo between ships and ports, performing maritime rescue, and offshore construction. Mechanically, marine cranes are similar to land-based cranes, both being underactuated systems with nonlinear and strongly coupled dynamic characteristics. However, because marine cranes are installed on ships, their operation is subject to continuous disturbances from the ship's six degrees of freedom motion (roll, pitch, yaw, and heave), leading to more severe load swaying problems compared to land-based cranes and significantly increasing control difficulty.

[0003] Existing technologies typically model marine crane dynamics in a geographic coordinate system (inertial frame). However, in practical engineering, sensors used to measure state variables such as slewing and swing angles are mounted on the crane itself, and their measurements are obtained in the ship's coordinate system (non-inertial frame). Therefore, these measurement signals must be converted to an inertial frame for model calculation and control. When a marine crane simultaneously exhibits five degrees of freedom motion (slewing, luffing, hoisting, and two-degree-of-freedom swing), some velocity variables in the inertial reference frame (such as slewing angular velocity and swing angular velocity) are difficult to measure directly. This coordinate transformation process not only introduces additional computational disturbances but may also cause singularity problems under specific attitudes, thus affecting the accuracy and stability of the control system. Furthermore, most existing studies assume that the crane is mounted on the ship's center of gravity during modeling, neglecting the offset between the crane's installation position and the ship's center of gravity. Simultaneously, marine cranes are subject to continuous matched and unmatched disturbances; effectively handling these disturbances to improve control robustness is also a pressing problem to be solved. Summary of the Invention

[0004] To address at least one of the technical problems mentioned above, this invention provides a control method and system for a five-degree-of-freedom marine crane in a non-inertial frame. The method involves establishing a five-degree-of-freedom dynamic model of the marine crane in a non-inertial frame and designing an adaptive radial basis function neural network based on this model to estimate and compensate for lumped disturbances online.

[0005] To achieve the above objectives, a first aspect of the present invention provides a control method for a five-degree-of-freedom marine crane in a non-inertial frame of reference, comprising:

[0006] The real-time state variables of the crane in a non-inertial frame are obtained, including the slewing angle, boom luffing angle, cable length, load surface interior angle, and load surface exterior angle.

[0007] The real-time state variables are input into an adaptive radial basis neural network. After online learning and compensation by the neural network, an estimate of the lumped disturbance is obtained. The neural network is designed based on a pre-built marine crane dynamic model. The dynamic model is a five-degree-of-freedom model established in the ship coordinate system using the Lagrange equation and the principle of virtual work, and considering the six-degree-of-freedom motion of the ship and the offset between the crane installation position and the ship's center of gravity.

[0008] Calculate the tracking error function based on the real-time state variables and the preset expected trajectory;

[0009] The estimated value of the lumped disturbance and the tracking error function are input to the feedback controller. After calculation by the feedback control law, the slewing control torque, the luffing control torque, and the cable control force are generated. The control torque and the force are output to the crane actuator to drive the crane to track the desired trajectory and suppress load sway.

[0010] Furthermore, the steps for constructing the dynamic model of the marine crane include:

[0011] Obtain the ship's roll rate, pitch rate, bow rate, and linear acceleration in the ship's coordinate system, as well as the offset vector between the crane's installation position and the ship's center of gravity;

[0012] Based on the angular velocity, linear acceleration, and offset vector, calculate the traction acceleration at the origin of the crane coordinate system, as well as the Coriolis acceleration of the boom and the load.

[0013] Based on the principle of virtual work, the generalized forces corresponding to the five degrees of freedom are calculated. The generalized forces include the control input torque, inertial force, Coriolis force and gravity term.

[0014] Substituting the generalized force into the second kind of Lagrange equation, and through derivation and merging, a five-degree-of-freedom dynamic model is obtained, represented by the inertia matrix, the centripetal-Coriolis matrix, the gravity vector, and the inertial force vector.

[0015] Furthermore, the step of inputting the real-time state variables into an adaptive radial basis neural network to obtain an estimate of the lumped perturbation also includes:

[0016] The real-time state variables are used as input to the neural network. Through the mapping and weighting of radial basis functions, the output value of the network is obtained as an estimate of the lumped perturbation.

[0017] Based on the tracking error function, the weights of the neural network are adjusted online according to a preset update law, wherein the update law is: ,in, For the defined error function, This is the output of the radial basis function.

[0018] Furthermore, the feedback control law is expressed as:

[0019]

[0020] in, The control input vector includes slewing control torque, luffing control torque, and cable control force. For the defined error function, The output of the adaptive radial basis function neural network is the estimated value of the lumped perturbation. The generalized inertia matrix after model transformation. and For the control gain matrix, It is a symbolic function.

[0021] Furthermore, the tracking error function is defined as follows:

[0022]

[0023]

[0024] in, For the first One state variable, For its expected value, =1,2,3,4,5 correspond to the slewing angle, boom luffing angle, cable length, load surface interior angle, and load surface exterior angle. It is a diagonal positive definite matrix. For the tracking error of the all-drive subsystem, For the tracking error of the rotation angle, For the tracking error of the boom amplitude, For tracking error of cable length, The derivative of the tracking error, For the tracking error of the underactuated subsystem, For the tracking error of the interior angle of the load surface, This represents the tracking error at the outer angle of the load surface.

[0025] Furthermore, it also includes: performing stability analysis on the closed-loop system based on Lyapunov functions, ensuring that the state variables of the fully driven subsystem asymptotically converge to the desired trajectory through the weight update law of the neural network, and ensuring that the state variables of the underdriven subsystem are bounded.

[0026] A second aspect of the present invention provides a control system for a five-degree-of-freedom marine crane in a non-inertial frame of reference, comprising:

[0027] The state acquisition module is used to acquire the real-time state variables of the crane in a non-inertial frame of reference. The state variables include slewing angle, boom luffing angle, cable length, load surface inner angle, and load surface outer angle.

[0028] The neural network estimation module is used to input the real-time state variables into the adaptive radial basis neural network. After the neural network learns and compensates online, it obtains the estimated value of the lumped disturbance. The neural network is designed based on a pre-built marine crane dynamic model. The dynamic model is a five-degree-of-freedom model established in the ship coordinate system using the Lagrange equation and the principle of virtual work, and considering the six-degree-of-freedom motion of the ship and the offset between the crane installation position and the ship's center of gravity.

[0029] The error calculation module is used to calculate the tracking error function based on the real-time state variables and the preset expected trajectory.

[0030] The feedback control module is used to input the estimated value of the lumped disturbance and the tracking error function to the feedback controller, and generate the slewing control torque, the luffing control torque and the cable control force through the feedback control law calculation; and output the control torque and the force to the crane actuator to drive the crane to track the desired trajectory and suppress load sway.

[0031] A third aspect of the present invention provides an electronic device including a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps of the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in the first aspect of the present invention.

[0032] A fourth aspect of the present invention provides a computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the steps of the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in the first aspect of the present invention.

[0033] A fifth aspect of the present invention provides a computer program product comprising software code, wherein the program in the software code performs the steps of the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in the first aspect of the present invention.

[0034] Compared with the prior art, the control method and system for a five-degree-of-freedom marine crane in a non-inertial frame of reference provided by the present invention have the following advantages:

[0035] (1) This invention establishes a five-degree-of-freedom dynamic model directly in a non-inertial frame (ship coordinate system) and takes into account the offset between the crane installation position and the ship's center of gravity. This avoids the state variable coordinate transformation process required for traditional inertial frame modeling, fundamentally eliminating the computational disturbance and singularity problems introduced by this, and improving the accuracy of the model and the stability of the control.

[0036] (2) The nonlinear feedback controller based on adaptive radial basis neural network (ARBFNN) designed in this invention can estimate and compensate for lumped disturbances composed of ship motion, friction, unmodeled dynamics, etc. online, which significantly enhances the robustness and disturbance rejection capability of the control system under complex sea conditions.

[0037] (3) The present invention proves the stability of the closed-loop system through rigorous Lyapunov theory, and verifies the superiority of the present invention over the traditional LQR controller in terms of trajectory tracking accuracy and load swing suppression through detailed comparative experiments. At the same time, the excellent robustness is fully demonstrated by introducing wind disturbance and irregular sea wave disturbance. Attached Figure Description

[0038] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0039] Figure 1 This is a flowchart of the control method for a five-degree-of-freedom marine crane in a non-inertial frame provided in Embodiment 1 of the present invention;

[0040] Figure 2 This is a schematic diagram of a marine crane provided in Embodiment 1 of the present invention;

[0041] Figure 3 This is a schematic diagram of the comparative experimental results provided in Embodiment 1 of the present invention (black solid line: expected value; blue solid line: controller effect proposed in the present invention; red dashed line: LQR control effect).

[0042] Figure 4 This is a schematic diagram of the ARBFNN output in the comparative experimental results provided in Embodiment 1 of the present invention;

[0043] Figure 5 This is a schematic diagram of the tracking error in the comparative experimental results provided in Embodiment 1 of the present invention;

[0044] Figure 6 This is a schematic diagram of the experimental results of robustness verification experiment 1 provided in Embodiment 1 of the present invention (black solid line: expected value; blue solid line: effect of the controller proposed in this invention);

[0045] Figure 7 This is a schematic diagram of the ARBFNN output in robustness verification experiment 1 provided in Embodiment 1 of the present invention;

[0046] Figure 8 This is a schematic diagram of the ship's roll and heave trajectories in the robustness verification experiment 2 provided in Embodiment 1 of the present invention;

[0047] Figure 9 This is a schematic diagram of the experimental results of robustness verification experiment 2 provided in Embodiment 1 of the present invention (black solid line: expected value; blue solid line: effect of the controller proposed in this invention);

[0048] Figure 10 This is a schematic diagram of the ARBFNN output in robustness verification experiment 2 provided in Embodiment 1 of the present invention;

[0049] Figure 11 This is an architecture diagram of the control system for a five-degree-of-freedom marine crane in a non-inertial frame, provided in Embodiment 2 of the present invention. Detailed Implementation

[0050] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0051] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form as well. Furthermore, it should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0052] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0053] All data acquisition in this embodiment is carried out in accordance with laws and regulations and with user consent, and the data is used legally.

[0054] Example 1

[0055] This embodiment provides a control method for a five-degree-of-freedom marine crane in a non-inertial frame. Its core lies in directly modeling and controlling the crane in the ship's coordinate system (a non-inertial frame), avoiding the problems caused by coordinate transformation. The following is a combination of... Figures 1 to 10 The method described in this embodiment will be explained in detail.

[0056] like Figure 1 As shown, the method provided in this embodiment includes the following steps:

[0057] S1. Obtain the real-time state variables of the crane in a non-inertial frame.

[0058] The state variables include the rotation angle. boom amplitude angle Cable length , Load surface interior angle and the outer corner of the load surface These variables are all obtained by direct measurement in the ship's coordinate system (non-inertial frame) by sensors (such as encoders) installed on the crane body. In this embodiment, since the model is directly built in the non-inertial frame, these measurements can be used directly for subsequent model calculations and control without complex coordinate transformations.

[0059] S2. The real-time state variables are input into an adaptive radial basis neural network. After online learning and compensation by the neural network, an estimate of the lumped disturbance is obtained. The neural network is designed based on a pre-built marine crane dynamic model. The dynamic model is a five-degree-of-freedom model established in the ship coordinate system using the Lagrange equation and the principle of virtual work, and considering the six-degree-of-freedom motion of the ship and the offset between the crane installation position and the ship's center of gravity.

[0060] The specific steps in constructing the pre-built marine crane dynamics model are as follows:

[0061] First, establish a coordinate system. For example... Figure 2 As shown, this method involves three coordinate systems: the geographic inertial coordinate system. Ship coordinate system (Non-inertial frame) and crane coordinate system Among them, the ship coordinate system The axis points forward (in the longitudinal direction of the ship). The axis is parallel to the deck and points laterally across the ship. The coordinate axes of the crane coordinate system and the ship coordinate system always coincide, and there is no relative motion between the two coordinate systems. However, the origins of the two coordinate systems do not coincide; there is an offset between them. The offset vector is defined as... This vector represents the offset between the crane's installation position and the ship's center of gravity. The ship has six degrees of freedom of motion, including roll, pitch, bow, and translation along three coordinate axes. The relationship between the ship's coordinate system and its inertial frame is determined by the rotation matrix. Translation vector describe.

[0062] Secondly, kinematic analysis is performed. The ship's roll, pitch, and bow angular velocities in the ship's coordinate system are obtained. angular acceleration and linear acceleration Based on this, the entrainment acceleration at the origin of the crane's coordinate system is calculated:

[0063] .

[0064] Let point A be any point on the boom, point B be the top of the boom, and point C be the load. In the crane coordinate system, the position vectors of points A and C are:

[0065] ,

[0066] in for Distance to point A This refers to the boom length. The length of the cable and its direction vector are given. and The expression is:

[0067]

[0068] .

[0069] Then the relative velocities of points A and C can be calculated. and absolute acceleration (including relative acceleration) , traction acceleration and Coriolis acceleration ).

[0070] Next, calculate the generalized forces. Based on the principle of virtual work, calculate the forces corresponding to the five degrees of freedom. generalized force First, the virtual displacement vectors of points C and A are obtained by taking partial derivatives, and then auxiliary vectors are introduced respectively. For the boom, the moment of inertia acting at point A is:

[0071] .

[0072] For load, inertial force Coriolis inertial force for:

[0073] .

[0074] Then, corresponding to the variable generalized force for:

[0075] ,

[0076] ,

[0077] ,

[0078] ,

[0079] .

[0080] in, To account for the equivalent gravitational acceleration after the ship moves, and These refer to the mass of the boom and the mass of the load, respectively.

[0081] Finally, a dynamic model is established. In a non-inertial frame of reference... Under, the kinetic energy of the load and the kinetic energy of the boom They are respectively:

[0082] ,

[0083] .

[0084] Kinetic energy of the entire crane system Define state variables. .

[0085] According to the second kind of Lagrange equation:

[0086] ,

[0087] After derivation and merging, the final five-degree-of-freedom dynamic model represented in matrix form is obtained:

[0088] ,

[0089] in The inertia matrix, For a centripetal-Coriolis matrix, To control the input vector, The gravity vector Let be the vector of inertial forces, and we have:

[0090]

[0091] .

[0092] In this invention, It also integrates friction. In addition, other unmodeled dynamics constitute the lumped disturbance to be estimated.

[0093] The model has the following properties:

[0094] Property 1. It is a positive definite matrix, that is ,in and It is a positive constant.

[0095] Property 2. For any vector ,have .

[0096] The control objective of this method can be stated as: transporting the load from the starting point to the target point in an inertial frame while suppressing load oscillation. In the inertial frame, the desired position of the load is... Therefore, according to the coordinate transformation relationship, we have:

[0097] According to the above formula, the expected value of the state variable is... It can be calculated. From 1 to 5. The calculation formula is:

[0098] ,

[0099] ,

[0100] ,

[0101] ,

[0102] ,

[0103] in , , , .

[0104] At this point, the control objective becomes the design control input. , and This causes the state variables to converge to their respective desired trajectories. That is:

[0105] .

[0106] Based on the above accurate model, in order to achieve the control objective, a controller needs to be designed to... To facilitate controller design, the model is decomposed into a complete drive subsystem ( ) and underactuated subsystem ( The dynamic equations can be rewritten as:

[0107] ,

[0108] ,

[0109] in ,

[0110] ,

[0111] ,

[0112] ,

[0113] .

[0114] Substituting the above equation into the transformation, we get:

[0115] ,

[0116] in , , , , .

[0117] Define complex nonlinear functions It contains linear acceleration that is difficult to measure. and angular acceleration Handling this function is quite challenging. To address this problem, this invention designs an Adaptive Radial Basis Function Neural Network (ARBFNN) for online estimation. .

[0118] Based on the RBF neural network approximation principle, the following lemma applies.

[0119] Lemma 1. Assumptions For definition in compact set A continuous function in the function. Then, for any constant... There exists an RBFNN whose output is .function and The differences between them can converge uniformly to a fairly small set. That is... .

[0120] According to Lemma 1, This can be represented by the following neural network:

[0121] .

[0122] in This is the ideal weight matrix. This is the input to the neural network. This is the activation function. This is the error.

[0123] Therefore, the output of RBFNN is: .

[0124] Define error for .

[0125] in This represents the weight estimation error.

[0126] The weights of the neural network are adjusted online according to the following update law:

[0127] ,

[0128] in For the error function defined later, This is the output of the radial basis function. Through this update law, the network can learn and compensate for changing lumped disturbances in real time.

[0129] S3. Calculate the tracking error function based on the real-time state variables and the preset expected trajectory.

[0130] First, define the tracking error:

[0131] .

[0132] Then, construct a filtering error function for the controller design:

[0133] ,

[0134] Where is a diagonal positive definite matrix. The introduction of this function simplifies controller design and guarantees that when... During convergence, the tracking error of the full-drive subsystem It will inevitably come to a halt.

[0135] S4. The estimated value of the lumped disturbance and the tracking error function The input is sent to the feedback controller, which calculates the feedback control law to generate slewing control torque, luffing control torque, and cable control force. The control torque and force are then output to the crane actuator to drive the crane to track the desired trajectory and suppress load sway.

[0136] Based on the model transformation in step S2 and the error function in step S3, and substituting them into the relevant expressions, we can obtain:

[0137] ,

[0138] Where auxiliary function .

[0139] Considering the passivity of the system (property 2), the following nonlinear feedback control law is designed:

[0140] ,

[0141] in, , The control gain matrix is ​​a diagonally positive definite matrix. Used to ensure robustness It is a sign function. The structure of this control law is clear: the first term... This is proportional-derivative feedback, used in sedation systems; the second term This is the feedforward compensation term, used to offset the lumped disturbance estimated by ARBFNN; the third term... Used to compensate for the time-varying inertia of the system; fourth item This is a robust term used to suppress errors in the neural network. This ensures the robustness of the system.

[0142] Theorem 1: For the expression The described marine crane system, in control law and the law of renewal Under its influence, it can ensure that the fully driven state variables asymptotically converge to the desired value or desired trajectory, and ensure the bounded input bounded output (BIBO) stability of the underdriven subsystem.

[0143] Proof: Define the Lyapunov function as follows:

[0144] .

[0145] Differentiate the function V and... Substituting the expression and the control law, we get:

[0146]

[0147] .

[0148] Considering ,and Items can be eliminated and substituted into the update law. (Right now ), we can obtain:

[0149] .

[0150] Using property 2 and the property of trace It can be eliminated Items, and merge The final result is:

[0151] .

[0152] because , can be obtained ,therefore:

[0153] .

[0154] This indicates Furthermore, define It can be proven According to the above formula and Given the boundedness of the property, and using Barbalat's lemma, we can obtain:

[0155] .

[0156] The above analysis proves the full drive subsystem asymptotic stability.

[0157] Next, we will analyze the state variables of the underactuated subsystem. and The stability of the dynamic model. Rewrite the dynamic model as about and Explicit form:

[0158]

[0159]

[0160] in , From these expressions, it can be seen that... and The dynamics are influenced by bounded inputs. These bounded inputs include ship motion parameters. , , and the state variables of the entire drive subsystem , , Since the fully driven subsystem has been proven to be asymptotically stable, these inputs are bounded. Therefore, the above system can be considered as bounded-input-bounded-output stable. Established.

[0161] Based on the above analysis, Theorem 1 is proved.

[0162] Experimental results verification

[0163] To verify the effectiveness and superiority of the method provided by this invention, in situations such as Figure 2 Comparative and robustness verification experiments were conducted on the self-made experimental platform shown. The platform consisted of a Stewart platform (simulating ship motion) and a slewing crane, with its luffing, slewing, and wire rope winch each driven by independent motors. The joint angles and the load swing angle were measured using a 17-bit incremental encoder. The Simulink compensation step size was set to 0.01 s, and the control board's maximum operating frequency was 500 Hz. The ship's roll and heave motions were generated using the MATLAB Marine System Simulator (MSS) toolbox.

[0164] The experimental platform parameters are as follows: load quality boom quality boom length .

[0165] It should be noted that this invention establishes a complete six-degree-of-freedom ship motion model. However, due to the limitations of the Stewart platform's motion range in the experimental platform, and based on existing research findings, only the ship's heave and pitch motions are simulated and experimentally verified. Existing research indicates that heave and pitch motions have the most significant impact on offshore lifting operations.

[0166] 1. Comparative Experiment

[0167] The controller proposed in this invention is compared with a linear quadratic regulator (LQR). After debugging, the LQR controller parameters are as follows:

[0168] ,

[0169] ,

[0170] .

[0171] The controller parameters of this invention are:

[0172] ,

[0173] ,

[0174] ,

[0175] RBF neural network parameters: center point ,width .

[0176] The ship's motion is set as regular pitching. and rise and fall .

[0177] Experimental results are as follows Figure 3 , Figure 4 , Figure 5 As shown. Figure 3 The response curves of five state variables are shown under the action of two controllers. As can be seen from the figure, under the action of both the LQR controller and the controller proposed in this invention, the system state variables converge to the desired value within approximately 5 seconds. Specifically, for the LQR controller, the rise times of the driven state variables are 2.27 seconds, 2.29 seconds, and 2.03 seconds, respectively. For the controller proposed in this invention, the rise times are 2.32 seconds, 3.70 seconds, and 1.76 seconds, respectively. Furthermore, neither controller exhibits significant overshoot.

[0178] However, significant differences exist between the two controllers in terms of tracking accuracy and sway suppression performance. To quantitatively compare their performance, the average tracking error and maximum residual sway angle error were calculated after the system entered the steady-state phase. The time for the system to enter the steady-state phase was... The calculation results are summarized in Table 1. The results show that the average tracking error of the controller of this invention is reduced by approximately 31.1% to 65% compared to the LQR controller. Furthermore, under the same initial swing angle error conditions, the controller proposed in this invention can suppress the residual swing angle to a smaller range more quickly than the LQR controller. These experimental results fully verify the superiority of the controller of this invention in terms of tracking performance and swing suppression.

[0179]

[0180] 2. Robustness verification experiment

[0181] Experiment 1: Continuous Wind Disturbance and Instantaneous Disturbance. Throughout the experiment, a continuous wind disturbance of 3.2 m / s was applied to the load using a fan, and an instantaneous external disturbance was applied to the load at t=35s. The experimental results are as follows: Figure 6 , Figure 7 As shown.

[0182] The results show that the controller of this invention can still effectively control the system under both continuous and transient disturbances. In particular, after being subjected to a transient impact at t=35s, resulting in a large swing of the load (in-plane angle amplitude of 11.3° and out-of-plane angle of 6.1°), the controller can suppress the residual swing back to within ±1° within approximately 21 seconds and 12 seconds, respectively, demonstrating strong anti-disturbance capability.

[0183] Experiment 2: Irregular Wave Disturbance. Using the MSS toolbox, irregular waves with different periods and amplitudes were generated to simulate the irregular rolling and heave motions of a ship, such as... Figure 8 As shown. The experimental results are as follows. Figure 9 , Figure 10 As shown.

[0184] The results show that under irregular wave disturbances, the desired trajectory of the system itself also exhibits irregular changes, but the controller of the present invention can still make the state variable track the desired value within about 6.4 seconds and effectively suppress residual oscillations. Figure 9 middle The brief amplification of tracking error that occurred at t=44.7s was due to the increased coupling effect caused by the increased rate of change of ship motion. However, the controller was still able to quickly correct the deviation within about 3.5 seconds, which once again verified its robustness.

[0185] In summary, the five-degree-of-freedom marine crane control method in non-inertial frame proposed in this embodiment effectively solves the disturbance and singularity problems caused by coordinate transformation in traditional methods through accurate non-inertial frame modeling and intelligent feedback control based on ARBFNN, and achieves high-precision trajectory tracking and robust load sway suppression.

[0186] Example 2

[0187] like Figure 11 This embodiment provides a control system for a five-degree-of-freedom marine crane in a non-inertial frame of reference, including:

[0188] The state acquisition module is used to acquire the real-time state variables of the crane in a non-inertial frame of reference. The state variables include slewing angle, boom luffing angle, cable length, load surface inner angle, and load surface outer angle.

[0189] The neural network estimation module is used to input the real-time state variables into the adaptive radial basis neural network. After the neural network learns and compensates online, it obtains the estimated value of the lumped disturbance. The neural network is designed based on a pre-built marine crane dynamic model. The dynamic model is a five-degree-of-freedom model established in the ship coordinate system using the Lagrange equation and the principle of virtual work, and considering the six-degree-of-freedom motion of the ship and the offset between the crane installation position and the ship's center of gravity.

[0190] The error calculation module is used to calculate the tracking error function based on the real-time state variables and the preset expected trajectory.

[0191] The feedback control module is used to input the estimated value of the lumped disturbance and the tracking error function to the feedback controller, and generate the slewing control torque, the luffing control torque and the cable control force through the feedback control law calculation; and output the control torque and the force to the crane actuator to drive the crane to track the desired trajectory and suppress load sway.

[0192] Example 3

[0193] Embodiment 3 of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements the steps in the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in Embodiment 1 of the present invention.

[0194] The detailed steps are the same as the control method for a five-degree-of-freedom marine crane in a non-inertial frame provided in Example 1, and will not be repeated here.

[0195] Example 4

[0196] Embodiment 4 of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in Embodiment 1 of the present invention.

[0197] The detailed steps are the same as the control method for a five-degree-of-freedom marine crane in a non-inertial frame provided in Example 1, and will not be repeated here.

[0198] The medium (auxiliary storage) for storing program code includes: hard disk drive, solid-state drive, electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), optical storage device, magnetic storage device, flash memory, magnetic disk or optical disk and / or combination of the above devices, and the main storage medium is random access memory (SROM).

[0199] Example 5

[0200] Embodiment 5 of the present invention provides a computer program product, including software code, wherein the program in the software code executes the steps in the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in Embodiment 1 of the present invention.

[0201] The detailed steps are the same as the control method for a five-degree-of-freedom marine crane in a non-inertial frame provided in Example 1, and will not be repeated here.

[0202] The computer product is stored on a data carrier and designed to execute the halophilic protein prediction method based on a hybrid deep learning architecture and a protein language model as described above. Therefore, the computer product according to this application has the advantage of providing a detailed description of the device with reference to this application. The computer program product can be encoded into computer-executable instruction code and can be executed by a suitable programming language such as Python. Furthermore, the computer program product can be provided on a network (e.g., the Internet), and Internet users can download the computer program product via the network (e.g., the Internet) when needed. The computer program product can be implemented using software, one or more dedicated electronic circuits (i.e., hardware), or a combination of software and hardware.

[0203] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0204] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0205] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0206] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0207] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A control method of a five-degree-of-freedom marine crane in a non-inertial system, characterized by, include: The real-time state variables of the crane in a non-inertial frame are obtained, including the slewing angle, boom luffing angle, cable length, load surface interior angle, and load surface exterior angle. The real-time state variables are input into an adaptive radial basis neural network. After online learning and compensation by the neural network, an estimate of the lumped disturbance is obtained. The neural network is designed based on a pre-built marine crane dynamic model. The dynamic model is a five-degree-of-freedom model established in the ship coordinate system using the Lagrange equation and the principle of virtual work, and considering the six-degree-of-freedom motion of the ship and the offset between the crane installation position and the ship's center of gravity. Calculate the tracking error function based on the real-time state variables and the preset expected trajectory; The estimated value of the lumped disturbance and the tracking error function are input to the feedback controller. After calculation by the feedback control law, the slewing control torque, the luffing control torque and the cable control force are generated. The control torque and the force are output to the crane actuator to drive the crane to track the desired trajectory and suppress load sway. The feedback control law is expressed as follows: in, The control input vector includes slewing control torque, luffing control torque, and cable control force. For the defined error function, The output of the adaptive radial basis function neural network is the estimated value of the lumped perturbation. The generalized inertia matrix after model transformation. and For the control gain matrix, It is a symbolic function; The tracking error function is defined as follows: in, For the first One state variable, For its expected value, =1,2,3 Corresponding to slewing angle, boom luffing angle, cable length, load surface interior angle, and load surface exterior angle, It is a diagonal positive definite matrix. For the tracking error of the all-drive subsystem, For the tracking error of the rotation angle, For the tracking error of the boom amplitude, For tracking error of cable length, The derivative of the tracking error, For the tracking error of the underactuated subsystem, For the tracking error of the interior angle of the load surface, This represents the tracking error at the outer angle of the load surface.

2. The method as described in claim 1, characterized in that, The steps for constructing the dynamic model of the marine crane include: Obtain the ship's roll rate, pitch rate, bow rate, and linear acceleration in the ship's coordinate system, as well as the offset vector between the crane's installation position and the ship's center of gravity; Based on the angular velocity, linear acceleration, and offset vector, calculate the traction acceleration at the origin of the crane coordinate system, as well as the Coriolis acceleration of the boom and the load. Based on the principle of virtual work, the generalized forces corresponding to the five degrees of freedom are calculated. The generalized forces include the control input torque, inertial force, Coriolis force and gravity term. Substituting the generalized force into the second kind of Lagrange equation, and through derivation and merging, a five-degree-of-freedom dynamic model is obtained, represented by the inertia matrix, the centripetal-Coriolis matrix, the gravity vector, and the inertial force vector.

3. The method as described in claim 1, characterized in that, The step of inputting real-time state variables into an adaptive radial basis neural network to obtain an estimate of the lumped perturbation also includes: The real-time state variables are used as input to the neural network. Through the mapping and weighting of radial basis functions, the output value of the network is obtained as an estimate of the lumped perturbation. Based on the tracking error function, the weights of the neural network are adjusted online according to a preset update law, wherein the update law is: ,in, For the defined error function, This is the output of the radial basis function.

4. The method as described in claim 1, characterized in that, Also includes: Stability analysis of the closed-loop system is performed based on Lyapunov functions. The weight update law of the neural network ensures that the state variables of the fully driven subsystem asymptotically converge to the desired trajectory, and ensures that the state variables of the underdriven subsystem are bounded.

5. A control system for a five-degree-of-freedom marine crane in a non-inertial frame of reference, characterized in that, include: The state acquisition module is used to acquire the real-time state variables of the crane in a non-inertial frame of reference. The state variables include slewing angle, boom luffing angle, cable length, load surface inner angle, and load surface outer angle. The neural network estimation module is used to input the real-time state variables into the adaptive radial basis neural network. After the neural network learns and compensates online, it obtains the estimated value of the lumped disturbance. The neural network is designed based on a pre-built marine crane dynamic model. The dynamic model is a five-degree-of-freedom model established in the ship coordinate system using the Lagrange equation and the principle of virtual work, and considering the six-degree-of-freedom motion of the ship and the offset between the crane installation position and the ship's center of gravity. The error calculation module is used to calculate the tracking error function based on the real-time state variables and the preset expected trajectory. The feedback control module is used to input the estimated value of the lumped disturbance and the tracking error function to the feedback controller, and generate the slewing control torque, the luffing control torque and the cable control force through the feedback control law calculation; and output the control torque and the force to the crane actuator to drive the crane to track the desired trajectory and suppress load sway. The feedback control law is expressed as follows: in, The control input vector includes slewing control torque, luffing control torque, and cable control force. For the defined error function, The output of the adaptive radial basis function neural network is the estimated value of the lumped perturbation. The generalized inertia matrix after model transformation. and For the control gain matrix, It is a symbolic function; The tracking error function is defined as follows: in, For the first One state variable, For its expected value, =1,2,3 Corresponding to slewing angle, boom luffing angle, cable length, load surface interior angle, and load surface exterior angle, It is a diagonal positive definite matrix. For the tracking error of the all-drive subsystem, For the tracking error of the rotation angle, For the tracking error of the boom amplitude, For tracking error of cable length, The derivative of the tracking error, For the tracking error of the underactuated subsystem, For the tracking error of the interior angle of the load surface, This represents the tracking error at the outer angle of the load surface.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in any one of claims 1 to 4.

8. A computer program product, comprising software code, characterized in that, The program in the software code executes the steps of the control method for a five-degree-of-freedom marine crane in a non-inertial frame as described in any one of claims 1 to 4.

Images