Control method and system applied to robot mechanical arm of converter station oil sample collection machine

Abstract

This invention provides a control method and system for a robotic arm used in oil sample collection at a converter station. The method includes: constructing a kinematic model, a dynamic model, and a three-stage cascaded motor system model; designing a non-matched composite disturbance filter; designing a switching preset finite-time time-varying convergence function, and constructing a first control signal generation function based on the switching preset finite-time time-varying convergence function; designing a state constraint function based on dynamic boundaries, and calculating dynamic boundary parameters using the pole placement method, then substituting the dynamic boundary parameters into the state constraint function to obtain the dynamic boundaries; performing expected convergence on the first control signal generation function based on the non-matched composite disturbance filter function and the dynamic boundaries to obtain a second control signal generation function, and obtaining the control signal for the robotic arm based on the second control signal generation function. This invention can meet the requirements of high-precision, stable, and safe oil sample collection operations in the complex environment of a converter station.

Description

Technical Field

[0001] This invention relates to the field of robot control technology, and in particular to a control method and system for a robotic arm used in oil sample collection at a converter station. Background Technology

[0002] As a critical facility in the power system, the condition of the insulating oil in converter stations (such as transformers and reactors) directly affects the safe operation of the system, necessitating regular oil sampling for testing. Traditional oil sampling relies on manual operation, which suffers from low efficiency, poor safety (in high-voltage environments), and insufficient sampling accuracy (due to human tremors). Therefore, the use of robotic arms to automate oil sampling has become a trend.

[0003] To meet the requirements of rapid response and time controllability in robotic arms, existing technologies propose methods such as finite-time control, fixed-time control, and preset-time control. Finite-time control utilizes strategies like the sliding surface of the end effector to achieve finite-time convergence, but the convergence time depends on the initial posture and parameters, making it difficult to guarantee time controllability when the initial state is variable. Fixed-time control introduces odd-order terms to make the convergence time independent of the initial state, but the estimated stabilization time is often too large and constrained by parameters, making it difficult to match industrial cycle time. Preset-time control allows for a preset upper limit on the convergence time, but the control input fluctuates drastically near the preset time, and robustness decreases after the preset time, making it difficult to meet the requirements of continuous operation. Regarding disturbance resistance, traditional disturbance observers rely on the assumption of "bounded disturbance derivatives," resulting in poor performance against sudden pulses and other disturbances; extended state observers have limited ability to handle mismatched disturbances; adaptive control and intelligent algorithms are prone to insufficient estimation accuracy or high computational complexity when dealing with complex disturbances. In terms of state constraint control, the obstacle Lyapunov function causes the control signal to jitter when the state approaches the boundary and is prone to breaking the constraint due to sudden disturbances; the preset performance control function has a fixed performance function and is prone to failure under strong interference; the model predictive control has a heavy computational burden and poor real-time performance, making it difficult to use in high-speed operation scenarios. Summary of the Invention

[0004] The purpose of this invention is to provide a control method and system for a robotic arm used in oil sampling at converter stations, aiming to solve problems such as insufficient controllability of control time, limited ability to suppress complex disturbances, and contradiction between state constraints and robustness in traditional technologies.

[0005] In a first aspect, the present invention provides a control method for a robotic arm used in oil sampling at a converter station, the method comprising:

[0006] The kinematic model, dynamic model and three-stage motor system model of the converter station oil sampling robot arm were constructed sequentially.

[0007] Design an unmatched composite disturbance filter based on the three-stage coupled motor system model described above;

[0008] Design a switching preset finite-time time-varying convergence function, and construct a first control signal generation function based on the switching preset finite-time time-varying convergence function;

[0009] Design a state constraint function based on dynamic boundaries, and calculate the dynamic boundary parameters using the pole placement method. Substitute the dynamic boundary parameters into the state constraint function to obtain the dynamic boundary.

[0010] The first control signal generation function is expected to converge based on the unmatched composite perturbation filter and dynamic boundary to obtain the second control signal generation function, and the control signal of the robotic arm is obtained based on the second control signal generation function.

[0011] In some embodiments, the step of sequentially constructing the kinematic model, dynamic model, and three-stage cascaded motor system model of the converter station oil sample collection robot arm includes:

[0012] Construct a kinematic model based on the following formula:

[0013] ;

[0014] in, It is a homogeneous transformation matrix. Let be the angle of the i-th joint of the robotic arm. Let be the torsion angle of the i-th link of the robotic arm, representing the angle between the z-axis (joint axis) in the coordinate system of two adjacent links. Let be the offset of the i-th link of the robotic arm, representing the distance between the x-axis in the coordinate system of two adjacent links;

[0015] Construct a dynamic model based on the following formula:

[0016] ;

[0017] in, Let be a vector of the joint angles of the link. The inertia matrix depends on the joint angle. It is the sum of the Coriolis force and the centrifugal force. and These are the friction force vector and the gravitational torque vector, respectively. For joint torque, For external force disturbances applied to the torque of the robotic arm, Let be the vector of the joint angular acceleration of the link, and be the second derivative of q with respect to time;

[0018] Construct a three-stage multi-stage motor system model based on the following formula:

[0019] ;

[0020] in, Let be the torque constant of the motor. For the motor armature current, This is the derivative of the armature current with respect to time, i.e., the rate of change of current. For armature inductance, Let be the back electromotive force constant of the motor. This is the derivative of the motor's rotation angle with respect to time, i.e., the motor's angular velocity. For input control voltage, External environmental interference with voltage This is the circuit resistance.

[0021] In some embodiments, the step of designing an unmatched composite disturbance filter based on the three-stage cascaded motor system model includes:

[0022] make , , , , The three-stage multistage motor system model then transforms into:

[0023] ;

[0024] Specifically, , , , , , , ;

[0025] in, Let be the angular velocity of the robotic arm joint. The angular acceleration of the robotic arm joints, This is the product of the motor torque constant and the inverse of the inertia matrix, reflecting the coupling relationship between the motor torque constant and the inertial characteristics of the robotic arm. It is the reciprocal of the armature inductance, reflecting the dynamic response characteristics of the motor's electrical system to current. This is a nonlinear force term related to joint angular velocity, which comprehensively reflects the combined effects of nonlinear factors such as Coriolis force, centrifugal force, friction, and gravity on joint movement. This is a damping term in the electrical system related to the armature current, affecting the dynamic changes of the current. For the equivalent disturbance term related to external forces, For the equivalent disturbance term related to voltage interference, The inverse of the robotic arm's inertia matrix. The inertia matrix, The terms represent the Coriolis force and centrifugal force, reflecting the nonlinear coupling characteristics between joint movements. The friction term describes the frictional resistance experienced by the robotic arm's joints during movement. This is a gravity term, generated by the gravity of each link in the robotic arm;

[0026] Design an unmatched composite perturbation filter based on the following formula:

[0027] ;

[0028] in, For the internal state of the observer Time derivative, Here is the gain matrix of the observer. For the internal state variables of the observer, A nonlinear function related to the state of the robotic arm. This is an estimate of the disturbance. For another gain matrix, For the internal state of the observer Time derivative, For another internal state variable of the observer, auxiliary To track and estimate disturbances more accurately. A nonlinear function related to the state of the robotic arm. As an intermediate variable, For the final disturbance compensation signal, For the gain matrix, the intermediate variables are... It is converted into a signal used to compensate for disturbances in the robotic arm.

[0029] In some embodiments, the step of designing a switching preset finite-time time-varying convergence function and constructing a first control signal generation function based on the switching preset finite-time time-varying convergence function includes:

[0030] Design a switching preset finite-time time-varying convergent function based on the following formula:

[0031] ;

[0032] in, For switching, a preset finite-time time-varying convergence function is used. For positive integers, For preset time, and ;

[0033] make ,when hour, ,and ;when hour, ;

[0034] in, To adjust the dynamic characteristics of the system, These are the core parameters of the system. For the internal state variables of the perturbation observer, The derivatives of the system's core parameters. This is the maximum allowed value.

[0035] In some embodiments, the method further includes:

[0036] exist The stage involves time-varying transformation and is defined. ,in, , ;

[0037] Construct the first Lyapunov candidate function Derive its derivative And using Young's inequality for processing, a virtual control signal is designed. ,in, , , All are positive constants, and I is the identity matrix;

[0038] Constructing a second Lyapunov candidate function ,according to The derivative is used to construct the first control signal generation function:

[0039] ;

[0040] in, , All are positive numbers;

[0041] in, These are the state variables after the first stage of time-varying transformation. These are the state variables after the second-stage time-varying transformation. It is 2+m. For 1+m, For subsequent virtual control inputs, m > 0 represents an integer. Here is the gain matrix. For control signals, The nonlinear characteristic function of the system, The derivative of the virtual control input for subsequent design. This is the coefficient matrix of the actual controller. This is the coefficient matrix of the virtual control signal. This is the gain matrix;

[0042] exist The time-varying transformation is simplified to the following stages: ,in, , Design virtual control signals ,in All are positive numbers. Therefore, the first control signal generation function is:

[0043] ;

[0044] in, All are positive numbers. Here is the gain matrix. Here is the gain matrix. It means a diagonal matrix.

[0045] In some embodiments, the step of designing the state constraint function based on dynamic boundaries includes:

[0046] Design the state constraint function based on the following formula:

[0047] ;

[0048] in, The output variable of the state constraint function quantifies the degree to which the system state (such as joint angles and angular velocities) constraints are satisfied. The third coefficient (a positive constant) of the state constraint function is increased. This can prevent the state from exceeding the constraint due to excessive acceleration. The second coefficient (a positive constant) of the state constraint function is increased. This can enhance the suppression of constraints that allow for rapid deviations from the state. The first coefficient (positive constant) of the state constraint function is increased. Can enhance the The inhibitory ability, accelerate The speed of convergence to the negative interval This refers to the disturbance compensation coefficient (positive number) in the state constraint function, which compensates for the influence of system disturbances on the state constraints. The deviation variable in the state constraint function describes the deviation between the actual state and the constraint boundary. Set the baseline threshold for the constant term (positive constant) of the state constraint function. For the output variables of the state constraint function, As the first intermediate variable of the state constraint function, by The first derivative is obtained by relating it to the state deviation, enabling the constraint function to accurately respond to the impact of the state deviation on the constraint and avoiding constraint lag. The second intermediate variable of the state constraint function is... The first derivative is obtained by relating it to the state deviation derivative, which further improves the response accuracy of the constraint function to the dynamic characteristics of the state and ensures that the state satisfies the constraints during the dynamic process.

[0049] In some embodiments, the step of calculating the dynamic boundary parameters using the pole placement method includes:

[0050] Define a time constant ,in It is selected based on actual needs. Ideal settling time;

[0051] Dominant pole Non-dominant poles and ,in, and All are constants not less than 5. As the dominant pole, It is a non-dominant pole. It is a non-dominant pole;

[0052] The true characteristic polynomial is obtained from the filter system matrix. This true characteristic polynomial is then equated with the expected characteristic polynomial to determine the true characteristic polynomial. , , The value of .

[0053] In some embodiments, the step of performing expected convergence on the first control signal generation function based on the unmatched composite perturbation filter and dynamic boundary to obtain a second control signal generation function, and obtaining the control signal of the robotic arm based on the second control signal generation function includes:

[0054] definition , ,in, For the desired joint angle, For virtual control input, This refers to the actual joint angle;

[0055] The second control signal generation function is obtained according to the following formula:

[0056] ;

[0057] in, Let k be the k-th derivative of the desired joint angle. The tracking error of the system or the rate of change of the internal state of the disturbance observer, The tracking error or disturbance of the system represents the internal state of the observer, and k is the order.

[0058] Secondly, the present invention provides a control system for a robotic arm used in oil sample collection at a converter station, the system comprising:

[0059] The model building module is used to sequentially build the kinematic model, dynamic model, and three-stage motor system model of the converter station oil sample collection robot arm.

[0060] The filter design module is used to design an unmatched composite disturbance filter based on the three-stage coupled motor system model.

[0061] The control signal generation module is used to design a switching preset finite-time time-varying convergence function and to construct a first control signal generation function based on the switching preset finite-time time-varying convergence function.

[0062] The dynamic boundary calculation module is used to design a state constraint function based on dynamic boundaries and calculate dynamic boundary parameters through the pole placement method. The dynamic boundary parameters are then substituted into the state constraint function to obtain the dynamic boundary.

[0063] The control signal output module is used to perform expected convergence of the first control signal generation function based on the unmatched composite disturbance filter and dynamic boundary to obtain the second control signal generation function, and to obtain the control signal of the robotic arm based on the second control signal generation function.

[0064] Thirdly, the present invention provides a storage medium that stores one or more programs, which, when executed by a processor, implement the above-described control method for a robotic arm used in oil sample collection at a converter station.

[0065] Fourthly, the present invention provides an electronic device, the electronic device comprising a memory and a processor, wherein:

[0066] The memory is used to store computer programs;

[0067] When the processor executes the computer program stored in the memory, it implements the above-described control method for the robotic arm used in oil sample collection at the converter station.

[0068] Compared with the prior art, the present invention has the following advantages:

[0069] 1. Synergistic optimization of time control accuracy and stability: This invention adopts a switching preset time time-varying convergence function, which makes the convergence time of the robotic arm completely controllable (unaffected by the initial state and parameters), and maintains the robustness of the system through a smooth switching mechanism after the preset time, thus solving the contradiction between "insufficient time accuracy" and "weak anti-disturbance ability in the later stage" of traditional preset time control.

[0070] 2. Precise suppression of compound disturbances: This invention designs an unmatched compound disturbance observer that does not rely on the assumption of "bounded disturbance derivatives". It can simultaneously estimate additive disturbances, state-coupled implicit disturbances and disturbances with unknown multiplication coefficients, and achieves asymptotic tracking by combining adaptive compensation technology, which significantly improves the ability to suppress disturbances in complex environments.

[0071] 3. Flexible constraints of dynamic boundaries: This invention introduces a dynamic boundary filter based on pole configuration, which enables the state constraint boundary to expand or converge in real time according to the disturbance. This ensures that the robot arm does not go out of bounds and avoids the problem of "constraint and robustness cannot be achieved at the same time" in traditional fixed boundary control, thereby improving the safety of the system under strong disturbances.

[0072] In summary, by integrating three core technologies—"preset time switching control," "non-matching disturbance observation," and "dynamic boundary constraints"—this invention achieves a synergistic improvement in time controllability, disturbance rejection accuracy, and constraint robustness compared to existing robotic arm control methods, making it more suitable for complex scenarios involving robots. Attached Figure Description

[0073] Figure 1 This is a flowchart of a control method for a robotic arm used in oil sample collection at a converter station, as proposed in an embodiment of the present invention.

[0074] Figure 2 A schematic diagram illustrating the effect of non-matching disturbance estimation for a robotic arm;

[0075] Figure 3 This is a schematic diagram of the adaptive law;

[0076] Figure 4 This is a schematic diagram of the joint angle tracking performance of the robotic arm;

[0077] Figure 5 This is a schematic diagram of the control system for a robotic arm used in oil sample collection at a converter station, as proposed in an embodiment of the present invention.

[0078] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0079] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, 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. Unless otherwise defined, the technical or scientific terms used herein should have the ordinary meaning understood by those skilled in the art. The terms "comprising" and similar expressions used herein mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, but does not exclude other elements or objects.

[0080] like Figure 1 As shown, an embodiment of the present invention proposes a control method for a robotic arm used in oil sample collection at a converter station. The method includes steps S101 to S105, wherein:

[0081] Step S101: Construct the kinematic model, dynamic model, and three-stage motor system model of the converter station oil sample collection robot arm in sequence;

[0082] It should be noted that in this step, the standard DH parameter method is used to perform kinematic modeling of the converter station oil sampling robot arm. Based on the structure of the robot arm, a coordinate system is established for each joint, the D-H parameters of each joint are determined, and then the kinematic model is constructed according to the following formula:

[0083] ;

[0084] in, It is a homogeneous transformation matrix. Let be the angle of the i-th joint of the robotic arm. Let be the torsion angle of the i-th link of the robotic arm, and let represent the angle between the z-axis of two adjacent links in the coordinate system. Let be the offset of the i-th link of the robotic arm, representing the distance between the x-axis in the coordinate systems of two adjacent links.

[0085] Furthermore, in some embodiments, the dynamic model of the robotic arm is derived based on the Newton-Euler principle. From the base end to the tool end, the velocity and acceleration at the center of mass of each link are determined sequentially using extrapolation, and the inertial force and torque acting at the center of mass are determined based on the Newton-Euler method. For example, the angular velocity vector of link i... ,in, ; angular acceleration vector of link i , Then, starting from the tool end to the base end, using the inversion method based on the conversion of force and torque, and the balance between the forces and torques at both joints and the inertial forces and torques at the center of mass, the components of the torques at each joint about the axis (i.e., joint torques) are determined sequentially. Finally, the dynamic equations of the robotic arm are obtained:

[0086] ;

[0087] in, Let be a vector of the joint angles of the link. The inertia matrix depends on the joint angle. It is the sum of the Coriolis force and the centrifugal force. and These are the friction force vector and the gravitational torque vector, respectively. For joint torque, For external force disturbances applied to the torque of the robotic arm, Let be the vector of the joint angular acceleration of the link, and be the second derivative of q with respect to time.

[0088] Considering the impact of motor voltage disturbances on the robotic arm's motion accuracy, a motor model is introduced to extend the robotic arm's dynamics model into a three-stage cascaded motor system model:

[0089] ;

[0090] in, Let be the torque constant of the motor. For the motor armature current, This is the derivative of the armature current with respect to time, i.e., the rate of change of current. For armature inductance, Let be the back electromotive force constant of the motor. This is the derivative of the motor's rotation angle with respect to time, i.e., the motor's angular velocity. For input control voltage, External environmental interference with voltage This is the circuit resistance.

[0091] Step S102: Design an unmatched composite disturbance filter based on the three-stage coupled motor system model;

[0092] It should be noted that, , , , , The three-stage multistage motor system model then transforms into:

[0093] ;

[0094] in, and It is an unknown constant. It is a known constant. , It is a known smooth function. and These represent unmatched interference and matched interference, respectively.

[0095] in, Let be the angular velocity of the robotic arm joint. The angular acceleration of the robotic arm joints, This is the product of the motor torque constant and the inverse of the inertia matrix, reflecting the coupling relationship between the motor torque constant and the inertial characteristics of the robotic arm. It is the reciprocal of the armature inductance, reflecting the dynamic response characteristics of the motor's electrical system to current. This is a nonlinear force term related to joint angular velocity, which comprehensively reflects the combined effects of nonlinear factors such as Coriolis force, centrifugal force, friction, and gravity on joint movement. This is a damping term in the electrical system related to the armature current, affecting the dynamic changes of the current. For the equivalent disturbance term related to external forces, For the equivalent disturbance term related to voltage interference, The inverse of the robotic arm's inertia matrix. The inertia matrix, The terms represent the Coriolis force and centrifugal force, reflecting the nonlinear coupling characteristics between joint movements. The friction term describes the frictional resistance experienced by the robotic arm's joints during movement. This is a gravity term, generated by the gravity of the links in the robotic arm.

[0096] To address the various complex disturbances encountered by the robotic arm in the converter station, a non-matched composite disturbance observer is designed. The non-matched composite disturbance is defined. ,in satisfy , It is a known constant vector. Immeasurable It is a known matrix. Given a function, It is a vector of unknown parameters, where m and x are unknown constants. Assume... As input, the system It is stable in the input-state dynamic sense, and the observability of (X, Y) is guaranteed. Since (X, Y) is observable, there exists a constant vector. , making Let this be the Herwitz matrix. Introduce a positive definite symmetric matrix. ,satisfy ,in Represents the m-order identity matrix. Designed for estimating... Filter:

[0097] ;

[0098] in, For the internal state of the observer Time derivative, Here is the gain matrix of the observer. For the internal state variables of the observer, A nonlinear function related to the state of the robotic arm. This is an estimate of the disturbance. For another gain matrix, For the internal state of the observer Time derivative, For another internal state variable of the observer, auxiliary To track and estimate disturbances more accurately. A nonlinear function related to the state of the robotic arm. As an intermediate variable, For the final disturbance compensation signal, For the gain matrix, the intermediate variables are... It is converted into a signal used to compensate for disturbances in the robotic arm.

[0099] This observer can estimate the mismatched composite disturbances in the robotic arm system in real time, providing an accurate basis for subsequent control compensation.

[0100] Step S103: Design a switching preset finite-time time-varying convergence function, and construct a first control signal generation function based on the switching preset finite-time time-varying convergence function;

[0101] In this step, to meet the strict time requirements of the converter station oil sampling task, a switching preset finite-time time-varying convergence function is designed:

[0102] ;

[0103] in, For switching, a preset finite-time time-varying convergence function is used. For positive integers, For preset time, and ;

[0104] make ,when hour, ,and ;when hour, ;

[0105] in, To adjust the dynamic characteristics of the system, These are the core parameters of the system. For the internal state variables of the perturbation observer, The derivatives of the system's core parameters. This is the maximum allowed value.

[0106] Furthermore, in some embodiments, a switching preset finite-time control method is designed in two time stages for robotic arm joint angle control. The stage involves time-varying transformation and is defined. ,in, , , For the virtual control inputs in the subsequent design, L1 = 2 + m, L2 = 1 + m, where m > 0 is an integer; then, the first Lyapunov candidate function is constructed. Derive its derivative And using Young's inequality for processing, a virtual control signal is designed. ,in, , , All are positive constants, and I is the identity matrix;

[0107] Constructing a second Lyapunov candidate function ,according to The derivative is used to construct the first control signal generation function:

[0108] ;

[0109] in, , All are positive numbers;

[0110] in, These are the state variables after the first stage of time-varying transformation. These are the state variables after the second-stage time-varying transformation. It is 2+m. For 1+m, For subsequent virtual control inputs, m > 0 represents an integer. Here is the gain matrix. For control signals, The nonlinear characteristic function of the system, The derivative of the virtual control input for subsequent design. This is the coefficient matrix of the actual controller. This is the coefficient matrix of the virtual control signal. This is the gain matrix.

[0111] exist The time-varying transformation is simplified to the following stages: ,in, , Design virtual control signals ,in All are positive numbers. Therefore, the first control signal generation function is:

[0112] ;

[0113] in, All are positive numbers. Here is the gain matrix. Here is the gain matrix. It means a diagonal matrix.

[0114] Stability analysis proves that the switching preset finite time control method can ensure that all states in the closed-loop system converge to the desired state within a preset finite time, and the convergence process does not depend on the initial conditions and control parameters. After a preset time T, the system can continue to maintain stability, avoiding the chattering phenomenon commonly found in traditional switching controllers.

[0115] Step S104: Design a state constraint function based on dynamic boundaries, and calculate the dynamic boundary parameters using the pole placement method, so as to substitute the dynamic boundary parameters into the state constraint function to obtain the dynamic boundary;

[0116] It should be noted that, to ensure the robotic arm does not collide with converter station equipment or cause other dangerous situations during oil sample collection, a state constraint control method based on dynamic boundaries is designed. The dynamic boundary function is defined and obtained through a third-order dynamic boundary filter, thus yielding the dynamic boundary:

[0117] ;

[0118] in, The output variable of the state constraint function quantifies the degree to which the system state constraints are satisfied. The third coefficient of the state constraint function is increased. This can prevent the state from exceeding the constraint due to excessive acceleration. The second coefficient of the state constraint function is increased. This can enhance the suppression of constraints that allow for rapid deviations from the state. The first coefficient of the state constraint function is increased. Can enhance the The inhibitory ability, accelerate The speed of convergence to the negative interval This is the disturbance compensation coefficient in the state constraint function, used to compensate for the impact of system disturbances on the state constraints. The deviation variable in the state constraint function describes the deviation between the actual state and the constraint boundary. For the constant term of the state constraint function, set the baseline threshold for the constraint function. For the output variables of the state constraint function, As the first intermediate variable of the state constraint function, by The first derivative is obtained by relating it to the state deviation, enabling the constraint function to accurately respond to the impact of the state deviation on the constraint and avoiding constraint lag. The second intermediate variable of the state constraint function is... The first derivative is obtained by relating it to the state deviation derivative, which further improves the response accuracy of the constraint function to the dynamic characteristics of the state and ensures that the state satisfies the constraints during the dynamic process.

[0119] Furthermore, in some embodiments, the process of calculating dynamic boundary parameters using the pole placement method is as follows:

[0120] Define a time constant ,in It is selected based on actual needs. Ideal settling time;

[0121] Dominant pole Non-dominant poles and ,in, and All are constants not less than 5. As the dominant pole, It is a non-dominant pole. It is a non-dominant pole;

[0122] The true characteristic polynomial is obtained from the filter system matrix. This true characteristic polynomial is then equated with the expected characteristic polynomial to determine the true characteristic polynomial. , , The value of .

[0123] Step S105: Based on the unmatched composite disturbance filter and dynamic boundary, the first control signal generation function is expected to converge to obtain the second control signal generation function, and the control signal of the robotic arm is obtained based on the second control signal generation function.

[0124] It should be noted that, based on the above dynamic boundary, a state constraint control law is designed and defined. , ,in, For the desired joint angle, For virtual control input, This refers to the actual joint angle.

[0125] The second control signal generation function (control law) is obtained using the following formula:

[0126] ;

[0127] in, Let k be the k-th derivative of the desired joint angle. The tracking error of the system or the rate of change of the internal state of the disturbance observer, The tracking error or disturbance of the system represents the internal state of the observer, and k is the order.

[0128] This control law ensures that tracking error is not compromised. This ensures that the robotic arm always moves within a safe dynamic boundary, effectively avoiding the risk of collision with converter station equipment, while also guaranteeing the system's stability and control precision.

[0129] By designing control strategies in the above four aspects, and comprehensively utilizing methods such as multi-joint robotic arm dynamics modeling, non-matched composite disturbance observation, preset finite-time switching control, and state constraint control based on dynamic boundaries, high-precision, high-stability, and high-safety control of the robotic arm for oil sampling in converter stations can be achieved, meeting the needs of oil sampling operations in complex environments of converter stations.

[0130] like Figure 2 As shown, the designed unmatched disturbance observer can accurately track the actual disturbance within approximately 0.1 seconds under three operating conditions: impulse disturbance, combined disturbance, and time-varying disturbance, with the estimation error rapidly decreasing to near zero. This indicates that the observer has excellent disturbance estimation performance for the system, capable of capturing the dynamic change trend of the disturbance in a very short time, providing a reliable basis for subsequent control compensation. Furthermore, regardless of the type of disturbance, the observer's output signal remains stable without drastic fluctuations or oscillations, further verifying the robustness and adaptability of the designed observer.

[0131] like Figure 3 As shown, under all three disturbance conditions, the adaptive law can quickly adjust to suitable parameter values ​​and stabilize within a short time, indicating that the adaptive mechanism can effectively cope with external disturbances and system uncertainties, achieving online estimation and real-time compensation for unknown disturbances. Furthermore, the adaptive law remains stable throughout the experiment without significant oscillations, fully demonstrating the adaptive capability and anti-interference performance of the proposed control strategy.

[0132] like Figure 4As shown, the tracking performance of the robotic arm is compared under two conditions: with and without combined disturbances. It is evident that the proposed control method exhibits excellent tracking capability and robustness in both scenarios. Specifically, regardless of the presence of combined disturbances, the joint angles of the robotic arm converge rapidly to the desired trajectory, and the tracking speed remains almost consistent. Simultaneously, the tracking error remains within a very small range, and the error curves in both cases almost overlap, indicating that the impact of external disturbances on the system's dynamic response is effectively suppressed.

[0133] In summary, the control method for a robotic arm used in oil sample collection at a converter station proposed in this invention has the following advantages:

[0134] This invention achieves significant technical results by integrating three core technologies: "preset time switching control," "non-matched disturbance observation," and "dynamic boundary constraints." In terms of time control, a switching-type preset time time-varying convergence function is employed, making the robotic arm's convergence time completely controllable and unaffected by initial state and parameters. After the preset time, smooth switching maintains system robustness, resolving the contradiction between insufficient time accuracy and weak later-stage disturbance resistance in traditional time control methods. This meets the requirements for precise and long-term stable time control in scenarios such as converter station oil sample collection. Regarding disturbance resistance, the designed non-matched composite disturbance observer does not rely on the "bounded derivative of disturbance" assumption and can simultaneously estimate multiple complex disturbances. Combined with adaptive compensation technology, it achieves asymptotic tracking. Experiments show that it can accurately track within 0.1s under impulse, composite, and time-varying disturbances, with estimation errors approaching zero, significantly improving disturbance resistance in complex environments.

[0135] Regarding state constraints, the dynamic boundary filter based on pole configuration allows the constraint boundaries to be adjusted in real time according to disturbances. This ensures that the robotic arm's movement does not exceed the limits (such as avoiding collisions with converter station equipment) and avoids signal jitter when approaching the boundary, as is common in traditional fixed boundary control. It also prevents constraint breaches when facing sudden disturbances, improving safety under strong interference. Simultaneously, precise kinematic and dynamic modeling provides a reliable foundation for the control strategy, comprehensively improving the robotic arm's time controllability, disturbance rejection accuracy, and constraint robustness. Experiments verify that it can quickly converge to the desired trajectory regardless of the presence or absence of complex disturbances, with minimal and stable tracking errors, making it suitable for the high-precision, stable, and safe oil sample collection requirements of the complex environment of the converter station.

[0136] like Figure 5 As shown, one embodiment of the present invention proposes a control system for a robotic arm used in oil sample collection at a converter station, the system comprising:

[0137] Model building module 10 is used to sequentially build the kinematic model, dynamic model and three-stage motor system model of the converter station oil sample collection robot arm;

[0138] Filter design module 20 is used to design an unmatched composite disturbance filter based on the three-stage cascaded motor system model;

[0139] The control signal generation module 30 is used to design a switching preset finite time time-varying convergence function and construct a first control signal generation function based on the switching preset finite time time-varying convergence function.

[0140] The dynamic boundary calculation module 40 is used to design a state constraint function based on dynamic boundaries and calculate dynamic boundary parameters through the pole placement method, so as to substitute the dynamic boundary parameters into the state constraint function to obtain the dynamic boundary.

[0141] The control signal output module 50 is used to perform expected convergence of the first control signal generation function based on the unmatched composite disturbance filter and dynamic boundary to obtain the second control signal generation function, and to obtain the control signal of the robotic arm based on the second control signal generation function.

[0142] In another aspect, the present invention also proposes a storage medium on which one or more programs are stored, which, when executed by a processor, implement the above-described control method for the robotic arm of an oil sample collection robot in a converter station.

[0143] In another aspect, the present invention also proposes an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to realize the above-mentioned control method for the robotic arm of the oil sample collection robot in the converter station.

[0144] Those skilled in the art will understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain stored, communicated, propagated, or transmitted programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0145] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0146] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0147] While embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations can be made to these embodiments. However, it should be understood that such modifications and variations fall within the scope and spirit of the invention as set forth in the claims. Furthermore, the invention described herein may have other embodiments and can be implemented or carried out in various ways.

Claims

1. A control method applied to a robot mechanical arm of a converter station oil sample collection machine, characterized in that, The method comprises: kinematics model, dynamics model and three-order cascade motor system model of the robot mechanical arm for oil sample collection in the converter station are sequentially constructed; The kinematics model is constructed according to the following formula: ； wherein, is a homogeneous transformation matrix, is an angle of the i-th joint of the robot arm, is a twist angle of the i-th link of the robot arm, representing an angle between z-axes of two adjacent link coordinate systems, is a offset of the i-th link of the robot arm, representing a distance between x-axes of two adjacent link coordinate systems; The dynamics model is constructed according to the following formula: ； wherein, is a vector of the joint angles of the links, is an inertia matrix dependent on the joint angles, is the sum of the Coriolis force and the centrifugal force, and are a friction force vector and a gravity torque vector, respectively, is a joint torque, is an external force disturbance applied on the mechanical arm torque, is a vector of the joint angle acceleration of the links, being the second derivative of q with respect to time; The three-order cascade motor system model is constructed according to the following formula: ； wherein, is the torque constant of the motor, is the motor armature current, is the derivative of the armature current with respect to time, i.e. the current rate of change, is the armature inductance, is the back EMF constant of the motor, is the derivative of the motor angle with respect to time, i.e. the motor angular velocity, is the input control voltage, is the external environment disturbance to the voltage, is the circuit resistance; A non-matching composite disturbance filter is designed according to the three-order cascade motor system model; A switching preset finite time time-varying convergence function is designed, and a first control signal generation function is constructed according to the switching preset finite time time-varying convergence function; A state constraint function based on a dynamic boundary is designed, and a dynamic boundary parameter is calculated through a pole placement method, so that the dynamic boundary parameter is substituted into the state constraint function to obtain the dynamic boundary; The first control signal generation function is expected to converge according to the non-matching composite disturbance filter and the dynamic boundary, to obtain a second control signal generation function, and a control signal of the mechanical arm is obtained according to the second control signal generation function.

2. The control method applied to the converter station oil sample collection robot mechanical arm according to claim 1, characterized in that, The step of designing a non-matching composite disturbance filter according to the three-order cascade motor system model comprises: Let , , , , The third-order cascaded motor system model is transformed into: ； In particular, , , , , , , ; wherein, is the angular velocity of the joint of the robot arm, is the angular acceleration of the joint of the robot arm, is the product of the motor torque constant and the inverse of the inertia matrix, reflecting the coupling relationship between the motor torque constant and the inertia characteristics of the robot arm, is the inverse of the armature inductance, reflecting the dynamic response characteristics of the motor electrical system to the current, is the nonlinear force term related to the joint angular velocity, is the electrical system damping term related to the armature current, affecting the dynamic change of the current, is the equivalent disturbance term related to the external force disturbance, is the equivalent disturbance term related to the voltage disturbance, is the inverse of the inertia matrix of the robot arm, is the inertia matrix, is the Coriolis force and centrifugal force term, embodying the coupling nonlinear characteristics between joint motions, is the friction force term, describing the frictional resistance when the joint of the robot arm moves, is the gravity term, generated by the gravity of each link of the robot arm; The non-matching composite disturbance filter is designed according to the following formula: ； wherein is a time derivative of an internal state of the observer, is a gain matrix of the observer, is an internal state variable of the observer, is a nonlinear function related to the state of the robot arm, is an estimate of the disturbance, is another gain matrix, is a time derivative of an internal state of the observer, is another internal state variable of the observer, assisting in tracking and estimating the disturbance, is a nonlinear function related to the state of the robot arm, is an intermediate variable, is a final disturbance compensation signal, is a gain matrix, converting the intermediate variable into a signal used for compensating the disturbance of the robot arm.​​​ 3. The control method for a robot manipulator for an oil sampling machine in a converter station according to claim 2, characterized in that, The step of designing a switching preset finite time time-varying convergence function and constructing a first control signal generation function according to the switching preset finite time time-varying convergence function comprises: The switching preset finite time time-varying convergence function is designed according to the following formula: ； wherein, is a switching function, is a positive constant, is a predetermined time, and ; Let when , , and ; when , ; wherein, is a core parameter of the system, is a core parameter of the system, is an internal state variable of the disturbance observer, is a derivative of a core parameter of the system, is a maximum allowable value.

4. The control method for a robot manipulator for an oil sampling machine in a converter station according to claim 3, characterized in that, The method further comprises: In Phase, time-varying transformation is performed, defining wherein, , ; Construct the first Lyapunov candidate function , derive its derivative , and process it by using Young's inequality to design the virtual control signal where ; , are all normal numbers, and I is the unit matrix; Construct a second Lyapunov candidate function , according to the derivative of the first control signal generation function: ； wherein , are all normal numbers; wherein is the state variable after the first phase time-varying transformation, is the state variable after the second phase time-varying transformation, is 2 + m, is 1 + m, is the virtual control input for the subsequent design, m > 0 is an integer, is a gain matrix, is a control signal, is a nonlinear characteristic function of the system, is the derivative of the virtual control input for the subsequent design, is a coefficient matrix of the actual controller, is a coefficient matrix of the virtual control signal, is a gain matrix; At stage, the time-varying transformation simplifies to where , ; the virtual control signal is designed where are all positive constants, the first control signal generating function is ； wherein are normal numbers, is a gain matrix, is a gain matrix, means a diagonal matrix.

5. The control method for a robot manipulator for an oil sampling machine in a converter station according to claim 4, characterized in that, The step of designing a state constraint function based on a dynamic boundary comprises: The state constraint function is designed according to the following formula: ； wherein is an output variable of the state constraint function, quantifying the satisfaction degree of the constraint on the system state, is a third coefficient of the state constraint function, is a second coefficient of the state constraint function, is a first coefficient of the state constraint function, is a disturbance compensation coefficient in the state constraint function, compensating the effect of system disturbance on the state constraint, is a deviation variable in the state constraint function, describing the deviation of the actual state from the constraint boundary, is a constant term of the state constraint function, setting a reference threshold of the constraint function, is an output variable of the state constraint function, is a first intermediate variable of the state constraint function, obtained by associating the first order derivative of the state constraint function with the state deviation, is a second intermediate variable of the state constraint function, obtained by associating the first order derivative of the state constraint function with the derivative of the state deviation.

6. The control method for a robot manipulator for an oil sampling machine in a converter station according to claim 5, characterized in that, The step of calculating a dynamic boundary parameter through a pole placement method comprises: defining a time constant wherein is selected according to the actual requirements the ideal regulation time of the dominant pole , non-dominant pole and wherein and are constants not less than 5, is a dominant pole, is a non-dominant pole, is a non-dominant pole; The real characteristic polynomial is obtained from the filter system matrix, and is set equal to the corresponding expected characteristic polynomial to determine the values of , , .

7. The control method for a robot manipulator for an oil sampling machine in a converter station according to claim 6, characterized in that, The step of obtaining a second control signal generation function according to the non-matching composite disturbance filter and the dynamic boundary, and making the first control signal generation function expected to converge, and obtaining a control signal of the mechanical arm according to the second control signal generation function comprises: Definitions , wherein, is the desired joint angle, is the virtual control input, is the actual joint angle; The second control signal generation function is obtained according to the following formula: ； wherein, is the kth derivative of the desired joint angle, is the rate of change of the tracking error or disturbance observer internal state of the system, is the tracking error or disturbance observer internal state of the system, and k is the order.

8. A control system applied to a converter station oil sample collection robot mechanical arm, used to implement the control method applied to the converter station oil sample collection robot mechanical arm according to any one of claims 1-7, characterized in that, The system comprises: A model construction module is configured to sequentially construct a kinematics model, a dynamics model and a three-order cascade motor system model of a robot mechanical arm for oil sample collection in the converter station; A filter design module is configured to design a non-matching composite disturbance filter according to the three-order cascade motor system model; A control signal generation module is configured to design a switching preset finite time time-varying convergence function, and construct a first control signal generation function according to the switching preset finite time time-varying convergence function; A dynamic boundary calculation module is configured to design a state constraint function based on a dynamic boundary, and calculate a dynamic boundary parameter through a pole placement method, so that the dynamic boundary parameter is substituted into the state constraint function to obtain the dynamic boundary; A control signal output module is configured to make the first control signal generation function expected to converge according to the non-matching composite disturbance filter and the dynamic boundary, to obtain a second control signal generation function, and obtain a control signal of the mechanical arm according to the second control signal generation function.

9. A storage medium, characterized by The storage medium stores one or more programs, which when executed by the processor implement the control method applied to the robot mechanical arm of the converter station oil sample collection machine as claimed in any one of claims 1-7.

Images