A three-stage swing bridge type trolley track planning and intelligent control method

Abstract

This invention provides a trajectory planning and intelligent control method for a three-stage swing-type overhead crane, relating to the technical field of intelligent overhead crane systems. The method includes: defining a planar coordinate system and determining the position coordinates of the crane's trolley, hook, frame, and load within the planar coordinate system; obtaining a dynamic model using the Lagrange equation based on the position coordinates; designing a controller based on the dynamic model; setting up a two-layer radial basis function neural network and determining the control force; and determining the overall control law based on the position coordinates, the controller, and the control force. According to this invention, intelligent anti-sway control of the three-stage swing system improves control quality, enhances the system's anti-interference capability, and improves the overall system safety level.

Description

Technical Field

[0001] This invention relates to the field of intelligent overhead crane systems, and in particular to a method for trajectory planning and intelligent control of a three-stage swing bridge type overhead crane. Background Technology

[0002] Overhead cranes are large pieces of equipment used to transport heavy loads from one location to another, and are widely used in various industrial sectors, including manufacturing plants, construction sites, seaports, and docks. With accelerating industrialization and the ever-increasing demand for higher operating efficiency, the performance requirements for overhead crane control systems are becoming increasingly stringent. One of the main technical challenges of overhead cranes is effectively suppressing load sway, which not only reduces transportation efficiency but also poses significant safety hazards. Therefore, studying the sway mechanism of overhead cranes and developing advanced control strategies has significant practical and engineering implications. In related technologies, existing research mostly focuses on anti-sway control in single-swing and double-swing modes. In actual heavy load transportation, a triple-swing mode is typically used, where the load is suspended from a gantry, the gantry is suspended from a hook, and the hook is connected to the hoisting mechanism. Compared to single-swing and double-swing modes, triple-swing overhead cranes exhibit a higher degree of underactuation, more complex dynamic behavior, and greater control challenges. Therefore, problems such as inaccurate trajectory planning and inaccurate anti-sway control exist, making online trajectory planning and anti-sway control difficult, resulting in low operating efficiency in the triple-swing mode.

[0003] The information disclosed in the background section of this application is intended only to enhance the understanding of the general background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0004] This invention provides a three-stage swing bridge type overhead crane trajectory planning and intelligent control method, which can solve the technical problems of online trajectory planning and anti-sway control that are difficult to achieve in related technologies.

[0005] According to a first aspect of the present invention, a method for trajectory planning and intelligent control of a three-stage swing-type overhead crane is provided, comprising:

[0006] Define a plane coordinate system and determine the position coordinates of the overhead crane's trolley, hook, gantry, and load in the plane coordinate system;

[0007] Based on the stated position coordinates, a dynamic model is obtained using the Lagrange equations;

[0008] Design a controller based on the aforementioned dynamic model;

[0009] Based on the position coordinates and the dynamic model, a two-layer radial basis function neural network is set up to determine the control force;

[0010] The overall control law is determined based on the position coordinates, the controller, and the control force.

[0011] According to the present invention, defining a planar coordinate system and determining the position coordinates of the overhead crane's trolley, hook, gantry, and load within the planar coordinate system includes:

[0012] According to the formula:

[0013]

[0014] Define the position coordinates of the overhead crane's trolley in the plane coordinate system. The position coordinates of the hook in the plane coordinate system The position coordinates of the hanger in the plane coordinate system and the position coordinates of the load in the plane coordinate system ,in, This refers to the lateral displacement of the trolley. This refers to the length of the hoisting rope. The length of the suspension cable. For hanging length, For hooks and The angle along the axial direction, For hangers and The angle along the axial direction, For load and Angle along the axial direction.

[0015] According to the present invention, a dynamic model is obtained, comprising:

[0016] Based on the aforementioned position coordinates, and the kinetic and potential energy of the crane's trolley, hook, frame, and load, determine the Lagrange function;

[0017] Based on the Lagrange function, a dynamic model is obtained.

[0018] According to the present invention, determining the Lagrange function includes:

[0019] According to the formula:

[0020]

[0021] Obtain the Lagrange function L, where T is the kinetic energy of the system and V is the potential energy of the system. , , m is the mass of the trolley. For the quality of the hook, For the quality of the hanger, Let F be the load mass and F be the motor driving force of the trolley.

[0022] According to the present invention, setting the acceleration function of the online trajectory includes:

[0023] Set anti-sway item ,in, , For the swing angle, It is an adjustable parameter;

[0024] According to the formula

[0025]

[0026]

[0027]

[0028] Define the acceleration function for the online trajectory, where, For the trajectory of the trolley displacement, The reference trajectory is t, where t is the current time. It is a time variable.

[0029] According to the present invention, a controller is designed, comprising:

[0030] Based on the aforementioned dynamic model, a matrix-form dynamic model is obtained. ,in, , , , , , ;

[0031] Construct composite signals based on the matrix form of the control model. Sum of error signals ,in, , , All are parameters to be designed;

[0032] Determine the finite-time sliding surface based on the error signal. ,in, , , All of these are adjustable controllable gains;

[0033] Determine the equivalent control terms based on the finite-time sliding surface. ;

[0034] Based on the finite-time sliding surface, determine the fixed-time double-power convergence law. ,in, , , , , All of these are adjustable controllable gains;

[0035] Based on the fixed-time double-power convergence law, determine the switching control term. ;

[0036] The control law is obtained based on the switching control term and the equivalent control term. ,in, This is the first control value for the motor driving force;

[0037] Design the control law based on the controller.

[0038] According to the present invention, a two-layer radial basis function neural network is configured, and a control force is determined, comprising:

[0039] Based on the stated position coordinates, set the input vector of the two-layer radial basis function neural network input layer. ;

[0040] Based on the input vector, determine the output of the first hidden layer of the two-layer radial basis function neural network. Where m is the number of output nodes in the first hidden layer. This represents the mean value of the output nodes in the first hidden layer. The standard deviation of the output nodes of the first hidden layer;

[0041] Based on the output of the first hidden layer, determine the output of the second hidden layer of the two-layer radial basis function neural network. Where n is the number of output nodes in the second hidden layer. This represents the mean value of the output nodes in the second hidden layer. The standard deviation of the output nodes of the second hidden layer;

[0042] Based on the output of the second hidden layer, determine the control force of the output layer of the two-layer radial basis function neural network. ,in, For adaptive weight vectors, It is the estimation error of the network. .

[0043] According to the present invention, the adaptive learning rate of the two-layer radial basis function neural network is ,in, It is a stable parameter that is greater than 0.

[0044] According to the present invention, determining the overall control law includes:

[0045] According to the formula:

[0046]

[0047] Determine the overall control law, where F is the motor driving force of the trolley.

[0048] According to a second aspect of the present invention, a computer-readable storage medium is provided having computer program instructions stored thereon, which, when executed by a processor, implement the three-stage swing bridge crane trajectory planning and intelligent control method.

[0049] According to an embodiment of the present invention, the three-stage swing bridge type overhead crane trajectory planning and intelligent control method can define a plane coordinate system and determine the position coordinates of the trolley, hook, gantry and load of the overhead crane in the plane coordinate system, thereby obtaining a dynamic model and designing a controller, then designing a two-layer radial basis function neural network and determining the control force, thereby determining the overall control law based on the position coordinates, the controller and the control force.

[0050] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Other features and aspects of the invention will become clearer from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0052] Figure 1 A schematic diagram of a three-stage swing bridge type overhead crane system according to an embodiment of the present invention is shown as an example;

[0053] Figure 2 A schematic diagram of a three-stage swing-type overhead crane trajectory planning and intelligent control method according to an embodiment of the present invention is shown as an example;

[0054] Figure 3 A first set of simulation results according to an embodiment of the present invention is shown as an example;

[0055] Figure 4 A second set of simulation results according to an embodiment of the present invention is shown as an example;

[0056] Figure 5 A third set of simulation results according to an embodiment of the present invention is shown as an example;

[0057] Figure 6 The fourth set of simulation results according to an embodiment of the present invention is shown as an example. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0060] Figure 1 A schematic diagram of a three-stage swing bridge type overhead crane system according to an embodiment of the present invention is shown as an example;

[0061] Figure 2 An exemplary schematic diagram of a three-stage swing bridge crane trajectory planning and intelligent control method according to an embodiment of the present invention is shown. The control method includes: defining a plane coordinate system and determining the position coordinates of the crane's trolley, hook, hanger, and load in the plane coordinate system.

[0062] Based on the stated position coordinates, a dynamic model is obtained using the Lagrange equations;

[0063] Design a controller based on the aforementioned dynamic model;

[0064] Based on the position coordinates and the dynamic model, a two-layer radial basis function neural network is set up to determine the control force;

[0065] The overall control law is determined based on the position coordinates, the controller, and the control force.

[0066] According to an embodiment of the present invention, the three-stage swing bridge type overhead crane trajectory planning and intelligent control method can define a plane coordinate system and determine the position coordinates of the trolley, hook, hanger and load of the overhead crane in the plane coordinate system, thereby obtaining a dynamic model and designing a controller. A two-layer radial basis function neural network is set in the controller and the control force is determined, thereby determining the overall control law based on the position coordinates, the controller and the control force.

[0067] Example 1:

[0068] According to an embodiment of the present invention, defining a planar coordinate system and determining the position coordinates of the trolley, hook, gantry, and load of the overhead crane in the planar coordinate system includes: determining the position coordinates of the trolley of the overhead crane in the planar coordinate system according to formula (1). The position coordinates of the hook in the plane coordinate system The position coordinates of the hanger in the plane coordinate system and the position coordinates of the load in the plane coordinate system ,

[0069] (1)

[0070] in, This refers to the lateral displacement of the trolley. This refers to the length of the hoisting rope. The length of the suspension cable. For hanging length, For hooks and The angle along the axial direction, For hangers and The angle along the axial direction, For load and Angle along the axial direction.

[0071] Example 2:

[0072] According to an embodiment of the present invention, a dynamic model is obtained based on the position coordinates and the mass of the trolley, hook, frame, and load of the overhead crane, including: determining the Lagrangian function based on the position coordinates and the kinetic and potential energy of the trolley, hook, frame, and load of the overhead crane; obtaining the dynamic model based on the Lagrangian function; and setting the acceleration function of the online trajectory based on the position coordinates, the mass of the trolley, hook, frame, and load of the overhead crane, and adjustable parameters.

[0073] Example 3:

[0074] According to an embodiment of the present invention, determining the Lagrange function based on the position coordinates and the kinetic and potential energy of the crane's trolley, hook, frame, and load includes: obtaining the Lagrange function L according to the formula:

[0075] (2)

[0076] Where T is the kinetic energy of the system and V is the potential energy of the system. , , m is the mass of the trolley. For the quality of the hook, For the quality of the hanger, Let F be the load mass and F be the motor driving force of the trolley.

[0077] According to an embodiment of the present invention, substituting T and V into formula (2), the nonlinear dynamic model of the three-pendulum system can be obtained as follows: , , , To simultaneously achieve the dual objectives of positioning and anti-swaying, an online motion trajectory is planned for the trolley's operation. In the three-sway system, the coupling relationship between the trolley, hook, frame, and load is shown in the aforementioned nonlinear dynamic model, providing a theoretical basis for online trajectory planning. Based on the approximate relationship near the equilibrium point, by merging the formulas from the nonlinear dynamic model, the following can be obtained: An anti-sway mechanism can be introduced into the reference trajectory for positioning, thereby accurately locating the load and maximally attenuating its sway. Here, assuming A represents an arbitrary variable that changes with time, then... This represents the first derivative of A with respect to time. This represents the second derivative of A with respect to time.

[0078] Example 4:

[0079] According to an embodiment of the present invention, based on the position coordinates and the mass of the crane's trolley, hook, gantry, and load, as well as adjustable parameters, the acceleration function of the online trajectory is set, including:

[0080] Set anti-sway item ,in, , For the swing angle, It is an adjustable parameter; the acceleration function of the linear trajectory is set according to formulas (3), (4), and (5).

[0081] (3)

[0082] (4)

[0083] (5)

[0084] in, For the trajectory of the trolley displacement, The reference trajectory is t, where t is the current time. It is a time variable.

[0085] According to an embodiment of the present invention, The first six terms in the expression can be represented as a quadratic form, which is easy to prove. If and only if and hour This holds true. Taking the first derivative of the above equation with respect to time, we can obtain... Furthermore, the trolley acceleration is combined with the anti-sway element, and with... By constructing an online acceleration reference trajectory in the form of [formula], the acceleration function of the online trajectory can be obtained.

[0086] According to an embodiment of the present invention, a dynamic model is obtained using the Lagrange equation. Although the online trajectory (5) can suppress load swaying, this open-loop control cannot resist external disturbances and impairs robustness. Therefore, a closed-loop controller must be designed to ensure that the trolley accurately tracks the planned online reference trajectory.

[0087] Example 5:

[0088] According to an embodiment of the present invention, designing a controller based on the dynamic model includes: obtaining a matrix-form dynamic model based on the dynamic model. (6), among which, , , , , , Based on the matrix form of the control model, construct the composite signal. (7) and error signal (8), among which, , , All parameters are to be designed; the finite-time sliding surface is determined based on the error signal. (9) Among them, , , All are adjustable control gains; the equivalent control term is determined based on the finite-time sliding surface. (10); Based on the finite-time sliding surface, determine the fixed-time double-power convergence law. (11), among which, , , , , All are adjustable control gains; the switching control term is determined based on the fixed-time double power convergence law. (12); Based on the switching control term and the equivalent control term, obtain the control law. (13), among which, The first control value for the motor driving force is given; a controller is designed based on the control law.

[0089] According to an embodiment of the present invention, the controller of system (6) is designed to ensure that the trolley runs accurately along the online trajectory planned in equation (11) while minimizing the potential swaying of the hook, hanger, and load. Therefore, the system is decomposed into a displacement subsystem and an angle subsystem. Achieving precise positioning and eliminating swaying simultaneously fully utilizes the dynamic coupling relationship between the trolley, hook, hanger, and load. Inspired by this, a composite signal coupling displacement and angular motion is first constructed to improve the transient performance of the system, thereby obtaining the composite signal. Secondly, the error signal is designed as The finite-time sliding surface is designed as The first derivative of the equation for a finite-time sliding surface can be obtained from the following equation: For a nominal system, ignoring disturbances and setting s=0, the following equivalent control term can be obtained: To improve the system's robustness to uncertainties and ensure convergence time, a fixed-time bipower convergence law is designed to reduce system jitter, which takes the form of: Since both are adjustable gain controls, the switching control item is: This leads to the control law. The Lyapunov function is constructed as follows: Taking the first derivative and substituting it into the fixed-time double-power convergence law, we obtain... Thus, a fixed convergence time can be obtained. .when ,have This is valid. Based on the finite-time sliding surface, we can conclude that... Again, select a Lyapunov candidate function related to the error signal. By calculating the first derivative of this function, it can be expressed as During the sliding process along the sliding surface, the error Theoretically, it can converge to zero in a finite time, and the convergence time satisfies the condition, therefore... The stability analysis of the sliding mode control method is now complete. The system tracking error can be resolved in a finite time. Since the system converges to zero, the control law designed based on the above controller can theoretically ensure that the system tracking error converges to zero within a fixed time.

[0090] Example 6:

[0091] According to an embodiment of the present invention, setting a two-layer radial basis function neural network based on the position coordinates and the dynamic model, and determining the control force, includes: setting the input vector of the input layer of the two-layer radial basis function neural network based on the position coordinates. Based on the input vector, determine the output of the first hidden layer of the two-layer radial basis function neural network. (14) where m is the number of output nodes in the first hidden layer. This represents the mean value of the output nodes in the first hidden layer. The standard deviation of the output nodes of the first hidden layer is used; based on the output of the first hidden layer, the output of the second hidden layer of the two-layer radial basis function neural network is determined. (15), where n is the number of output nodes in the second hidden layer. This represents the mean value of the output nodes in the second hidden layer. The standard deviation of the output nodes of the second hidden layer is used; based on the output of the second hidden layer, the control force of the output layer of the two-layer radial basis function neural network is determined. (16), among which, For adaptive weight vectors, It is the estimation error of the network. .

[0092] According to an embodiment of the present invention, theoretically, the system error can converge to zero within a finite time, meaning the trolley will run accurately along the planned trajectory. However, in practical applications, unavoidable external uncertainties may reduce the robustness of the system. While increasing the controller gain is a common method to improve robustness, it also amplifies system jitter. Therefore, based on the proposed sliding mode variable structure control, a neural network is used to estimate and compensate for uncertainties, thereby mitigating jitter and improving robustness. To improve estimation accuracy, a two-layer radial basis function neural network with two hidden layers is employed. Compared to a single-layer radial basis function neural network, the two-layer architecture achieves higher estimation accuracy through multi-layer recombination and combination of activation functions and linear weighting. The two-layer radial basis function neural network consists of four parts: an input layer, a first hidden layer, a second hidden layer, and an output layer. A Gaussian function is used as the activation function for each node in the two hidden layers. When the number of nodes in the input layer is set to 2, the input vector can be represented as... The first hidden layer primarily maps the input signal from the input layer to the higher-dimensional hidden layer. It consists of a set of hidden neurons, each activated by a Gaussian function. For the... The output of a node can be represented as The main function of the second hidden layer is to map the output signal of the first hidden layer to the second hidden layer and apply the Gaussian function again. For the first hidden layer... The output of a node can be represented as .

[0093] According to an embodiment of the present invention, the output This represents the output of each neuron in the second hidden layer and its corresponding connection weight. The weighted sum, one of which has This allows the control force output by the output layer of a two-layer radial basis function neural network to be controlled. Defined as Furthermore, let's assume... ,in, , ,therefore, .

[0094] Example 7:

[0095] According to an embodiment of the present invention, the adaptive learning rate of the two-layer radial basis function neural network is... ,in, It is a stable parameter that is greater than 0.

[0096] Example 8:

[0097] Based on the position coordinates, the controller, and the control force, the overall control law is determined, including: determining the overall control law according to formula (17).

[0098] (17)

[0099] Where F represents the driving force of the trolley's motor. The closed-loop system is asymptotically stable.

[0100] According to an embodiment of the present invention, the three-stage swing bridge type overhead crane trajectory planning and intelligent control method can define a plane coordinate system and determine the position coordinates of the trolley, hook, gantry and load of the overhead crane in the plane coordinate system, thereby obtaining a dynamic model and designing a controller, setting a two-layer radial basis function neural network, and determining the control force, thereby determining the overall control law based on the position coordinates, the controller and the control force.

[0101] Example 9:

[0102] Figure 3 A first set of simulation results according to an embodiment of the present invention is shown as an example.

[0103] like Figure 3 As shown, in this simulation, four reference trajectories are selected: a three-segment acceleration trajectory, an acceleration trajectory embedding a ramp function, an acceleration trajectory embedding a trigonometric function, and the proposed online reference trajectory. A PD controller is selected to track these four reference trajectories. The target displacement of the trolley in the simulation is selected as... . Figure 3 The simulation results are presented. Figure 3 As can be seen, when using a PD controller to track four tracks, the offline track is detrimental to load transportation. The trolley reaches a steady state in 16.11 seconds with an overshoot of 0.41m. The maximum swing angles of the hook, hanger, and load are 4.81˚, 4.96˚, and 5.18˚, respectively. In contrast, the trolley moves smoothly along the planned online track, reaching the desired position in 10.76 seconds without overshoot. The maximum swing angles of the hook, hanger, and load are 2.6˚, 2.66˚, and 2.76˚, respectively, representing reductions of 46%, 46%, and 47%. Figure 3 The vertical motion curve of the payload during lifting operations was plotted, showing significant fluctuations and rebounds even after the trolley came to a standstill. This improvement stems from the feedback signal of the time-varying oscillation angle in the online trajectory.

[0104] Example 10:

[0105] Figure 4 A second set of simulation results according to an embodiment of the present invention is shown as an example.

[0106] like Figure 4 As shown, the proposed sliding mode variable structure controller is compared with the selected PD controller to verify the superiority of the proposed control method. After repeated debugging, the optimal control gain is obtained as follows: , , , , , , , The neural network structure is 2-7-5-1. Figure 4 The subplots, from top to bottom, depict the trolley displacement, hook, hanger, load swing angle, and control input. It is clear that the two controllers exhibit almost identical swing angles when tracking the online trajectory, remaining within 3˚.

[0107] Example 11:

[0108] Figure 5 A third set of simulation results according to an embodiment of the present invention is shown as an example.

[0109] like Figure 5 As shown, to verify the robustness of the proposed control method, a non-zero initial swing angle ( and The following section compares it with the PD method. Figure 5 Simulation results are presented. The gains of both control methods are consistent with the previous settings. Controlling the triple pendulum system presents certain challenges due to the strong coupling between its system states. Under initial angle excitation, the PD controller struggles to suppress the triple pendulum oscillations, while the proposed control method effectively mitigates the load oscillations and almost eliminates residual oscillations.

[0110] Example 12:

[0111] Figure 6 The fourth set of simulation results according to an embodiment of the present invention is shown as an example.

[0112] like Figure 6 As shown, to evaluate the anti-interference performance of the proposed method, a sinusoidal disturbance with an amplitude of 0.15 m was applied to the trolley track within a time period of 10-12 seconds. The results were compared with those of the PD method. Figure 6Simulation results are presented. Both the PD method and the proposed method can maintain the swing angles of the hook, hanger, and load within a reasonable range under disturbance conditions. However, the proposed method achieves a smaller swing amplitude and suppresses oscillations more quickly. This improvement is attributed to the use of an enhanced switching control law combined with a two-layer radial basis function neural network, thereby reducing the required control input. In contrast, the PD method requires a larger driving force to suppress external disturbances, which may lead to actuator saturation control failure. Therefore, the proposed method exhibits superior anti-disturbance performance under disturbance conditions.

[0113] According to an embodiment of the present invention, a computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions, when executed by a processor, implement the three-stage swing bridge crane trajectory planning and intelligent control method.

[0114] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0115] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments, and any variations or modifications may be made to the implementation of the present invention without departing from the stated principles.

Claims

1. A method for trajectory planning and intelligent control of a three-stage swing-type overhead crane, characterized in that, include: Define a plane coordinate system and determine the position coordinates of the overhead crane's trolley, hook, gantry, and load in the plane coordinate system; Based on the stated position coordinates, a dynamic model is obtained using the Lagrange equations; Design a controller based on the aforementioned dynamic model; Based on the position coordinates and the dynamic model, a two-layer radial basis function neural network is set up to determine the control force; The overall control law is determined based on the position coordinates, the controller, and the control force.

2. The three-stage swing bridge crane trajectory planning and intelligent control method according to claim 1, characterized in that, Define a plane coordinate system and determine the position coordinates of the overhead crane's trolley, hook, gantry, and load within that plane coordinate system, including: According to the formula: Define the position coordinates of the overhead crane's trolley in the plane coordinate system. The position coordinates of the hook in the plane coordinate system The position coordinates of the hanger in the plane coordinate system and the position coordinates of the load in the plane coordinate system ,in, This refers to the lateral displacement of the trolley. This refers to the length of the hoisting rope. The length of the suspension cable. For hanging length, For hooks and The angle along the axial direction, For hangers and The angle along the axial direction, For load and Angle along the axial direction.

3. The three-stage swing bridge type overhead crane trajectory planning and intelligent control method according to claim 2, characterized in that, Based on the stated position coordinates, a dynamic model is obtained using the Lagrange equations, including: Based on the position coordinates, as well as the displacement and velocity components of the overhead crane's trolley, hook, hanger, and load, determine the system's kinetic and potential energy. The Lagrange function is determined based on the position coordinates and the kinetic and potential energy of the crane's trolley, hook, frame, and load. Based on the Lagrange function, a dynamic model is obtained.

4. The three-stage swing bridge type overhead crane trajectory planning and intelligent control method according to claim 3, characterized in that, Based on the stated position coordinates, and the kinetic and potential energies of the crane's trolley, hook, support, and load, the Lagrangian function is determined, including: According to the formula: The Lagrange function L is obtained, where T is the kinetic energy of the system and V is the potential energy of the system. , , m is the mass of the trolley. For the quality of the hook, For the quality of the hanger, Let F be the load mass and F be the motor driving force of the trolley.

5. The three-stage swing bridge crane trajectory planning and intelligent control method according to claim 3, characterized in that, Based on the stated position coordinates and the mass of the overhead crane's trolley, hook, support, and load, as well as adjustable parameters, the acceleration function of the online trajectory is set, including: Set anti-sway item ,in, , For the swing angle, It is an adjustable parameter; According to the formula , , Define the acceleration function of the online trajectory, where, For the trajectory of the trolley displacement, The reference trajectory is t, where t is the current time. It is a time variable.

6. The three-stage swing bridge type overhead crane trajectory planning and intelligent control method according to claim 5, characterized in that, Based on the aforementioned dynamic model, a controller is designed, including: Based on the aforementioned dynamic model, a matrix-form dynamic model is obtained. ,in, , , , , , Construct composite signals based on the matrix-form dynamic model. Sum of error signals ,in, , , All are parameters to be designed; Determine the finite-time sliding surface based on the error signal. ,in, , , All are adjustable control gains; the equivalent control term is determined based on the finite-time sliding surface. Based on the finite-time sliding surface, determine the fixed-time double-power convergence law. ,in, , , , , All are adjustable control gains; the switching control term is determined based on the fixed-time double power convergence law. The control law is obtained based on the switching control term and the equivalent control term. ,in, The first control value for the motor driving force is given; based on the controller, the control law is designed.

7. The three-stage swing bridge crane trajectory planning and intelligent control method according to claim 6, characterized in that, Based on the position coordinates and the dynamic model, a two-layer radial basis function neural network is set up to determine the control force, including: Based on the stated position coordinates, set the input vector of the two-layer radial basis function neural network input layer. Based on the input vector, determine the output of the first hidden layer of the two-layer radial basis function neural network. Where m is the number of output nodes in the first hidden layer. This represents the mean value of the output nodes in the first hidden layer. The standard deviation of the output nodes of the first hidden layer is used; based on the output of the first hidden layer, the output of the second hidden layer of the two-layer radial basis function neural network is determined. Where n is the number of output nodes in the second hidden layer. This represents the mean value of the output nodes in the second hidden layer. The standard deviation of the output nodes of the second hidden layer is used; based on the output of the second hidden layer, the control force of the output layer of the two-layer radial basis function neural network is determined. ,in, For adaptive weight vectors, It is the estimation error of the network. .

8. The three-stage swing bridge crane trajectory planning and intelligent control method according to claim 7, characterized in that, The adaptive learning rate of a two-layer radial basis function neural network is ,in, It is an adjustable parameter that is greater than 0.

9. The three-stage swing bridge type overhead crane trajectory planning and intelligent control method according to claim 7, characterized in that, Based on the position coordinates, the controller, and the control force, the overall control law is determined, including: According to the formula: Determine the overall control law, where F is the motor driving force of the trolley.

10. A computer-readable storage medium, characterized in that, It stores computer program instructions that, when executed by a processor, implement the method of any one of claims 1-9.

Images