A galvanometer driving precision stability control method and system based on current closed loop

Abstract

The application discloses a galvanometer driving precision stability control method and system based on a current closed loop, relates to the technical field of galvanometer driving precision control, and comprises the following steps: obtaining a target position trajectory of a galvanometer, calculating a target angular velocity and a target angular acceleration; generating a torque feedforward compensation quantity and a voltage feedforward compensation quantity; collecting an actual position of the galvanometer, estimating an actual velocity and a disturbance state, generating a feedback correction quantity according to a position deviation, and compensating and correcting by using the disturbance state; fusing the feedback correction quantity and the torque feedforward compensation quantity to generate a current loop instruction, constructing a current closed loop control, generating a voltage correction quantity, fusing the voltage correction quantity with the voltage feedforward compensation quantity to form a driving control instruction, and driving the galvanometer to execute deflection; and simultaneously updating a state observer to generate a disturbance compensation quantity of a next control period. Through the cooperative action of feedforward compensation, disturbance observation compensation and current closed loop control, the galvanometer dynamic tracking precision, disturbance resistance and long-term operation stability are improved.

Description

Technical Field

[0001] This invention relates to the field of galvanometer drive precision control technology, specifically to a galvanometer drive precision stability control method and system based on current closed-loop control. Background Technology

[0002] A galvanometer is a precision actuator that converts electrical signals into mirror angle deflection. It is widely used in laser marking, laser cutting, laser welding, 3D printing, machine vision, and optical scanning. In practical applications, the control system outputs a drive signal to the galvanometer motor according to the target trajectory, causing the mirror to deflect at high speed along a predetermined path, thereby achieving precise positioning and scanning of the laser beam.

[0003] Existing galvanometer control systems typically employ a closed-loop position control structure, using a PID controller to generate drive commands based on position errors to achieve trajectory tracking. While this control method is simple in structure, it relies heavily on errors for control. When the target trajectory changes rapidly, the controller must compensate for the error only after it has occurred, resulting in a certain dynamic lag in the system. Furthermore, because the inertial torque, damping torque, and elastic restoring torque in the galvanometer's dynamic model are not compensated in advance, the burden on the feedback controller is easily increased during high-speed motion, affecting the system's response speed and control accuracy. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides a current-closed-loop-based method and system for stabilizing the driving accuracy of a galvanometer. This solution overcomes the limitations of existing galvanometer control systems, which typically employ a position closed-loop control structure and use a PID controller to generate drive commands based on position errors to achieve trajectory tracking. While this control method is simple in structure, it relies heavily on errors for control. When the target trajectory changes rapidly, the controller must compensate for the error after it has occurred, resulting in a certain dynamic lag in the system. Furthermore, because the inertial torque, damping torque, and elastic restoring torque in the galvanometer dynamics model are not compensated in advance, the burden on the feedback controller is easily increased during high-speed motion, affecting the system's response speed and control accuracy.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method and system for stabilizing the accuracy of a galvanometer drive based on a current closed-loop control, comprising: Obtain the target position trajectory of the galvanometer, and calculate the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamic model; The torque feedforward compensation amount is generated based on the target position, target angular velocity and target angular acceleration, and the voltage feedforward compensation amount is generated based on the galvanometer motor voltage model; The actual position of the galvanometer is acquired, and the actual velocity and disturbance state are estimated through the state observer. The disturbance state includes unmodeled friction disturbance, parameter perturbation, and external torque disturbance. A feedback correction amount is generated based on the positional deviation between the target position and the actual position, and a disturbance compensation amount is generated using the disturbance state to correct the feedback correction amount. The corrected feedback correction amount is combined with the torque feedforward compensation amount to obtain the current loop command; The actual current of the galvanometer motor is obtained, and a voltage correction amount is generated by the current loop controller based on the current deviation between the current loop command and the actual current. The voltage correction amount and the voltage feedforward compensation amount are combined to obtain the galvanometer motor drive control command; The galvanometer is driven to deflect according to the drive control command, and the state observer state is updated to enter the next control cycle.

[0006] Preferably, the step of acquiring the target position trajectory of the galvanometer, and calculating the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamic model, specifically includes: Acquire target position trajectory data sent by the upper control system, wherein the target position trajectory is a discrete time sequence sampled at a fixed control period; Within each control cycle, the target position of the current cycle and the target position of the previous cycle are read. The target angular velocity is calculated by first-order backward difference, and the target angular acceleration is calculated by second-order difference based on the target angular velocity of the current cycle and the target angular velocity of the previous cycle. The results of the differential operation are low-pass filtered to suppress high-frequency quantization noise introduced by the differential operation; When the target trajectory is a known analytical function, the target angular velocity and target angular acceleration can be obtained directly by differentiating the analytical expression. The calculated target position, target angular velocity, and target angular acceleration are used as input parameters for the galvanometer feedforward control.

[0007] Preferably, the torque feedforward compensation amount is generated based on the target position, target angular velocity, and target angular acceleration, specifically including: The rotational inertia, damping coefficient, and elastic coefficient of the galvanometer were obtained through an offline system identification experiment. The offline system identification experiment included applying a sweep frequency current excitation signal to the galvanometer, synchronously collecting deflection angle response data, and using the least squares method to fit the second-order transfer function for parameter identification. The torque constant of the galvanometer was determined independently using a constant current deflection experiment; A galvanometer dynamic model is established based on the target position, target angular velocity, and target angular acceleration. The galvanometer dynamic model is a second-order inertial-damped-elastic system model. Based on the galvanometer dynamics model, and in conjunction with the moment of inertia, damping coefficient, elastic coefficient, and torque constant, the inertial torque compensation term, damping torque compensation term, and elastic restoring torque compensation term are calculated respectively to obtain the torque feedforward compensation amount.

[0008] Preferably, the generation of voltage feedforward compensation based on the galvanometer motor voltage model specifically includes: Obtain the resistance, inductance, and back electromotive force parameters of the galvanometer motor; Based on the rate of change of current command in adjacent cycles, calculate the dynamic inductance compensation amount used to overcome electrical inertia when the current changes rapidly. Based on the current command of the previous cycle, calculate the resistor voltage drop compensation amount to avoid current build-up lag caused by resistor voltage division. Based on the target angular velocity, calculate the back electromotive force compensation amount used to pre-counteract the induced electromotive force generated by the coil motion; The voltage feedforward compensation is obtained by superimposing at least one of the inductance dynamic compensation, resistance voltage drop compensation, and back electromotive force compensation.

[0009] Preferably, the acquisition of the actual position of the galvanometer and the estimation of the actual velocity and disturbance state through a state observer specifically includes: The actual position information of the galvanometer is acquired by a position sensor, and the acquired position signal is filtered. The actual location information is input into the extended state observer ESO, which is a third-order linear extended state observer LESO. The extended state observer outputs three state variables: estimated position, estimated actual velocity, and total disturbance state. The estimated actual velocity is reconstructed by the extended state observer based on the system model and position measurement information, rather than obtained by directly performing differential operations on the actual position. The total disturbance state estimates the unmodeled frictional disturbance, parameter perturbation, and external torque disturbance as a unified extended state variable. The total disturbance state is converted into an equivalent disturbance compensation current, which is used as the disturbance compensation amount to correct the feedback correction amount.

[0010] Preferably, the process of driving the galvanometer to deflect according to the drive control command and updating the state observer state specifically includes: The torque feedforward compensation amount and the corrected feedback correction amount are weighted and fused to generate a current loop command, wherein the feedforward weight and the feedback weight are dynamically adjusted according to the operating conditions, and the weight switching adopts a linear gradual transition. The coil current is detected by a sampling resistor and a differential amplifier, and the actual current is obtained after analog-to-digital conversion. Calculate the current deviation between the current loop command and the actual current; The current deviation is adjusted in a closed loop by a PI controller to generate a corresponding voltage correction amount. The PI controller is equipped with a conditional integral anti-saturation mechanism. The voltage correction amount and the voltage feedforward compensation amount are superimposed to generate a drive control command; The drive control command is modulated by PWM to generate a drive signal, and the voltage is output to the galvanometer coil through the power circuit to drive the lens to deflect. The internal state variables of the state observer are updated using the actual position, actual current, and current loop command obtained in the current control cycle. The disturbance compensation amount for the next control cycle is generated based on the updated disturbance state.

[0011] A current-closed-loop-based galvanometer driving accuracy stabilization control system, used to implement the above-mentioned galvanometer driving accuracy stabilization control method, characterized in that it includes: The trajectory feedforward generation module is used to acquire the target trajectory of the galvanometer and calculate the target angular velocity and target angular acceleration. It generates torque feedforward compensation based on the galvanometer dynamics model and voltage feedforward compensation based on the galvanometer motor voltage model. The observation compensation module is used to acquire the actual position of the galvanometer and estimate the actual velocity and disturbance state through the state observer, generate feedback correction amount based on the position deviation and use the disturbance state for correction; The current closed-loop drive module is used to fuse the corrected feedback correction amount with the torque feedforward compensation amount to generate a current loop command, and generate a voltage correction amount based on the current deviation and fuse it with the voltage feedforward compensation amount to generate a drive control command to drive the galvanometer to perform deflection. The state iteration module is used to update the internal state variables of the state observer and generate the disturbance compensation amount for the next control cycle, thereby realizing the closed-loop iteration of the control cycle.

[0012] To address the aforementioned technical problems, this paper provides a current-closed-loop-based method and system for stabilizing the driving accuracy of a galvanometer. This solution overcomes the limitations of existing galvanometer control systems, which typically employ a position closed-loop control structure and use a PID controller to generate drive commands based on position errors to achieve trajectory tracking. While this control method is simple in structure, it relies heavily on errors for control. When the target trajectory changes rapidly, the controller must compensate for the error after it has occurred, resulting in a certain dynamic lag in the system. Furthermore, because the inertial torque, damping torque, and elastic restoring torque in the galvanometer dynamics model are not compensated in advance, the burden on the feedback controller is easily increased during high-speed motion, affecting the system's response speed and control accuracy.

[0013] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method and system for stabilizing the accuracy of a galvanometer drive based on a current closed-loop control, comprising: Obtain the target position trajectory of the galvanometer, and calculate the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamic model; The torque feedforward compensation amount is generated based on the target position, target angular velocity and target angular acceleration, and the voltage feedforward compensation amount is generated based on the galvanometer motor voltage model; The actual position of the galvanometer is acquired, and the actual velocity and disturbance state are estimated through the state observer. The disturbance state includes unmodeled friction disturbance, parameter perturbation, and external torque disturbance. A feedback correction amount is generated based on the positional deviation between the target position and the actual position, and a disturbance compensation amount is generated using the disturbance state to correct the feedback correction amount. The corrected feedback correction amount is combined with the torque feedforward compensation amount to obtain the current loop command; The actual current of the galvanometer motor is obtained, and a voltage correction amount is generated by the current loop controller based on the current deviation between the current loop command and the actual current. The voltage correction amount and the voltage feedforward compensation amount are combined to obtain the galvanometer motor drive control command; The galvanometer is driven to deflect according to the drive control command, and the state observer state is updated to enter the next control cycle.

[0014] Preferably, the step of acquiring the target position trajectory of the galvanometer, and calculating the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamic model, specifically includes: Acquire target position trajectory data sent by the upper control system, wherein the target position trajectory is a discrete time sequence sampled at a fixed control period; Within each control cycle, the target position of the current cycle and the target position of the previous cycle are read. The target angular velocity is calculated by first-order backward difference, and the target angular acceleration is calculated by second-order difference based on the target angular velocity of the current cycle and the target angular velocity of the previous cycle. The results of the differential operation are low-pass filtered to suppress high-frequency quantization noise introduced by the differential operation; When the target trajectory is a known analytical function, the target angular velocity and target angular acceleration can be obtained directly by differentiating the analytical expression. The calculated target position, target angular velocity, and target angular acceleration are used as input parameters for the galvanometer feedforward control.

[0015] Preferably, the torque feedforward compensation amount is generated based on the target position, target angular velocity, and target angular acceleration, specifically including: The rotational inertia, damping coefficient, and elastic coefficient of the galvanometer were obtained through an offline system identification experiment. The offline system identification experiment included applying a sweep frequency current excitation signal to the galvanometer, synchronously collecting deflection angle response data, and using the least squares method to fit the second-order transfer function for parameter identification. The torque constant of the galvanometer was determined independently using a constant current deflection experiment; A galvanometer dynamic model is established based on the target position, target angular velocity, and target angular acceleration. The galvanometer dynamic model is a second-order inertial-damped-elastic system model. Based on the galvanometer dynamics model, and in conjunction with the moment of inertia, damping coefficient, elastic coefficient, and torque constant, the inertial torque compensation term, damping torque compensation term, and elastic restoring torque compensation term are calculated respectively to obtain the torque feedforward compensation amount.

[0016] Preferably, the generation of voltage feedforward compensation based on the galvanometer motor voltage model specifically includes: Obtain the resistance, inductance, and back electromotive force parameters of the galvanometer motor; Based on the rate of change of current command in adjacent cycles, calculate the dynamic inductance compensation amount used to overcome electrical inertia when the current changes rapidly. Based on the current command of the previous cycle, calculate the resistor voltage drop compensation amount to avoid current build-up lag caused by resistor voltage division. Based on the target angular velocity, calculate the back electromotive force compensation amount used to pre-counteract the induced electromotive force generated by the coil motion; The voltage feedforward compensation is obtained by superimposing at least one of the inductance dynamic compensation, resistance voltage drop compensation, and back electromotive force compensation.

[0017] Preferably, the acquisition of the actual position of the galvanometer and the estimation of the actual velocity and disturbance state through a state observer specifically includes: The actual position information of the galvanometer is acquired by a position sensor, and the acquired position signal is filtered. The actual location information is input into the extended state observer ESO, which is a third-order linear extended state observer LESO. The extended state observer outputs three state variables: estimated position, estimated actual velocity, and total disturbance state. The estimated actual velocity is reconstructed by the extended state observer based on the system model and position measurement information, rather than obtained by directly performing differential operations on the actual position. The total disturbance state estimates the unmodeled frictional disturbance, parameter perturbation, and external torque disturbance as a unified extended state variable. The total disturbance state is converted into an equivalent disturbance compensation current, which is used as the disturbance compensation amount to correct the feedback correction amount.

[0018] Preferably, the process of driving the galvanometer to deflect according to the drive control command and updating the state observer state specifically includes: The torque feedforward compensation amount and the corrected feedback correction amount are weighted and fused to generate a current loop command, wherein the feedforward weight and the feedback weight are dynamically adjusted according to the operating conditions, and the weight switching adopts a linear gradual transition. The coil current is detected by a sampling resistor and a differential amplifier, and the actual current is obtained after analog-to-digital conversion. Calculate the current deviation between the current loop command and the actual current; The current deviation is adjusted in a closed loop by a PI controller to generate a corresponding voltage correction amount. The PI controller is equipped with a conditional integral anti-saturation mechanism. The voltage correction amount and the voltage feedforward compensation amount are superimposed to generate a drive control command; The drive control command is modulated by PWM to generate a drive signal, and the voltage is output to the galvanometer coil through the power circuit to drive the lens to deflect. The internal state variables of the state observer are updated using the actual position, actual current, and current loop command obtained in the current control cycle. The disturbance compensation amount for the next control cycle is generated based on the updated disturbance state.

[0019] A current-closed-loop-based galvanometer driving accuracy stabilization control system, used to implement the above-mentioned galvanometer driving accuracy stabilization control method, characterized in that it includes: The trajectory feedforward generation module is used to acquire the target trajectory of the galvanometer and calculate the target angular velocity and target angular acceleration. It generates torque feedforward compensation based on the galvanometer dynamics model and voltage feedforward compensation based on the galvanometer motor voltage model. The observation compensation module is used to acquire the actual position of the galvanometer and estimate the actual velocity and disturbance state through the state observer, generate feedback correction amount based on the position deviation and use the disturbance state for correction; The current closed-loop drive module is used to fuse the corrected feedback correction amount with the torque feedforward compensation amount to generate a current loop command, and generate a voltage correction amount based on the current deviation and fuse it with the voltage feedforward compensation amount to generate a drive control command to drive the galvanometer to perform deflection. The state iteration module is used to update the internal state variables of the state observer and generate the disturbance compensation amount for the next control cycle, thereby realizing the closed-loop iteration of the control cycle.

[0020] The drive output unit is used to superimpose the voltage correction amount and the voltage feedforward compensation amount to obtain the drive control command, and generate a drive signal through PWM modulation to control the power circuit to output voltage to the galvanometer coil, thereby driving the lens to deflect. The state update unit is used to update the internal state variables of the extended state observer using the actual position, actual current and current loop command obtained in the current control cycle. The disturbance iteration unit is used to retain the updated total disturbance state to the next control cycle and convert it into a disturbance compensation amount; The loop triggering unit is used to trigger the next control cycle after the state update is completed, and return to the trajectory feedforward generation module to continue execution, forming a closed loop iteration.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows: In this invention, a galvanometer accuracy and stability control architecture is constructed, consisting of trajectory feedforward compensation, extended state observer disturbance compensation, and current closed-loop control. Through the fusion control of torque feedforward compensation, voltage feedforward compensation, disturbance compensation, and current closed-loop feedback, joint compensation for the mechanical dynamic characteristics of the galvanometer, the electrical characteristics of the motor, and external disturbances is achieved, forming a high-precision control link from the target trajectory to the actual output torque, which significantly improves the dynamic tracking accuracy and control stability of the galvanometer system.

[0022] This invention introduces an extended state observer to observe the galvanometer's operating state in real time, enabling online estimation of the actual speed and total disturbance state without the need for additional speed sensors. It also unifies unmodeled frictional disturbances, parameter perturbations, and external torque disturbances into a unified total disturbance state for compensation, thereby enhancing the system's adaptability to complex operating conditions and unknown disturbances, and improving control accuracy and anti-disturbance performance.

[0023] In this invention, a trajectory feedforward generation mechanism is constructed to calculate the target angular velocity and target angular acceleration based on the target position trajectory. The torque feedforward compensation amount and voltage feedforward compensation amount are generated by combining the galvanometer dynamic model and the motor voltage model, respectively. The main dynamic characteristics of the galvanometer mechanical system and electrical system are pre-compensated before the control command is executed, reducing the adjustment burden of the position loop and current loop, and improving the galvanometer's response speed and tracking capability to high-speed scanning trajectories. Attached Figure Description

[0024] Figure 1 This is a flowchart of the galvanometer driving accuracy stabilization control method in this invention; Figure 2 This is a flowchart illustrating the generation of torque feedforward and voltage feedforward in this invention; Figure 3 This is a flowchart of the extended state observation and disturbance compensation process in this invention; Figure 4 This is a framework diagram of the galvanometer driving accuracy stability control system of the present invention. Detailed Implementation

[0025] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0026] Reference Figure 1-3 As shown, a method for stabilizing the accuracy of a galvanometer drive based on a current closed-loop control specifically includes: Obtain the target position trajectory of the galvanometer, and calculate the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamic model; The torque feedforward compensation amount is generated based on the target position, target angular velocity and target angular acceleration, and the voltage feedforward compensation amount is generated based on the galvanometer motor voltage model; The actual position of the galvanometer is acquired, and the actual velocity and disturbance state are estimated through the state observer. The disturbance state includes unmodeled friction disturbance, parameter perturbation, and external torque disturbance. A feedback correction amount is generated based on the positional deviation between the target position and the actual position, and a disturbance compensation amount is generated using the disturbance state to correct the feedback correction amount. The corrected feedback correction amount is combined with the torque feedforward compensation amount to obtain the current loop command; The actual current of the galvanometer motor is obtained, and a voltage correction amount is generated by the current loop controller based on the current deviation between the current loop command and the actual current. The voltage correction amount and the voltage feedforward compensation amount are combined to obtain the galvanometer motor drive control command; The galvanometer is driven to deflect according to the drive control command, and the state observer state is updated to enter the next control cycle.

[0027] Furthermore, the acquisition of the target position trajectory of the galvanometer, and the calculation of the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamic model, specifically includes: Acquire target position trajectory data sent by the upper control system, wherein the target position trajectory is a discrete time sequence sampled at a fixed control period; Within each control cycle, the target position of the current cycle and the target position of the previous cycle are read. The target angular velocity is calculated by first-order backward difference, and the target angular acceleration is calculated by second-order difference based on the target angular velocity of the current cycle and the target angular velocity of the previous cycle. The results of the differential operation are low-pass filtered to suppress high-frequency quantization noise introduced by the differential operation; When the target trajectory is a known analytical function, the target angular velocity and target angular acceleration can be obtained directly by differentiating the analytical expression. The calculated target position, target angular velocity, and target angular acceleration are used as input parameters for the galvanometer feedforward control. In this scheme, the target position trajectory data is sent to the controller according to the processing graphic. The target position trajectory is a discrete time series and the time interval between adjacent data points is equal to the control cycle. In each control cycle, the current target position and the target position of the previous cycle are read. The target angular velocity is calculated by first-order backward difference, and then the target angular velocity is differentially obtained by second-order difference. To suppress the high-frequency quantization noise introduced by the differential operation, a first-order low-pass filter is connected after the differential operation; For known analytical trajectories such as sinusoidal scanning, the target angular velocity and target angular acceleration can be obtained directly by differentiating the analytical expressions, without the need for numerical difference. The final calculated current target position, target angular velocity, and target angular acceleration are used as input parameters for the galvanometer feedforward control, and are called by the torque feedforward and voltage feedforward circuits.

[0028] Furthermore, a torque feedforward compensation amount is generated based on the target position, target angular velocity, and target angular acceleration, specifically including: The rotational inertia, damping coefficient, and elastic coefficient of the galvanometer were obtained through an offline system identification experiment. The offline system identification experiment included applying a sweep frequency current excitation signal to the galvanometer, synchronously collecting deflection angle response data, and using the least squares method to fit the second-order transfer function for parameter identification. The torque constant of the galvanometer was determined independently using a constant current deflection experiment; A galvanometer dynamic model is established based on the target position, target angular velocity, and target angular acceleration. The galvanometer dynamic model is a second-order inertial-damped-elastic system model. Based on the galvanometer dynamics model, and in conjunction with the moment of inertia, damping coefficient, elastic coefficient, and torque constant, the inertial torque compensation term, damping torque compensation term, and elastic restoring torque compensation term are calculated respectively to obtain the torque feedforward compensation amount; In this scheme, the mechanical dynamics of the galvanometer are described using a second-order inertial-damped-elastic system model, satisfying: ; in, For rotational inertia, The damping coefficient is... The elastic coefficient, It is the torque constant. , , These are the angular acceleration, angular velocity, and angle of the galvanometer, respectively. The parameters mentioned above, which are used to drive the motor current, were obtained through offline system identification experiments and stored in the controller parameter table. Angular acceleration, angular velocity, and angle of the galvanometer The parameters mentioned above, which are used to drive the motor current, were obtained through offline system identification experiments and stored in the controller parameter table. The specific identification process involves applying a 0-500Hz sweep frequency current excitation signal to the galvanometer, simultaneously acquiring deflection angle response data, fitting the second-order transfer function using the least squares method, and identifying the moment of inertia, damping coefficient, and elastic coefficient. The torque constant was determined separately through a constant current deflection experiment. Torque feedforward compensation Calculate in real time using the following formula: ; in: These are the target angular acceleration, target angular velocity, and target position, respectively. Inertial torque compensation term Damping torque compensation term used to overcome inertial drag torque during acceleration. The elastic restoring torque compensation term is used to overcome the viscous damping torque proportional to velocity. Used to overcome the spring restoring torque that is proportional to the angle; The torque feedforward compensation provides the main driving torque component required for the galvanometer to track the target trajectory, significantly reducing the dynamic burden on the position feedback loop.

[0029] Furthermore, the generation of voltage feedforward compensation based on the galvanometer motor voltage model specifically includes: Obtain the resistance, inductance, and back electromotive force parameters of the galvanometer motor; Based on the rate of change of current command in adjacent cycles, calculate the dynamic inductance compensation amount used to overcome electrical inertia when the current changes rapidly. Based on the current command of the previous cycle, calculate the resistor voltage drop compensation amount to avoid current build-up lag caused by resistor voltage division. Based on the target angular velocity, calculate the back electromotive force compensation amount used to pre-counteract the induced electromotive force generated by the coil motion; The voltage feedforward compensation is obtained by superimposing at least one of the inductance dynamic compensation, resistance voltage drop compensation and back electromotive force compensation. In this scheme, the electrical characteristics of the galvanometer motor coil satisfy the voltage equation: ; in, For coil resistance, For inductance, The back electromotive force coefficient, This is the motor terminal voltage. This is the motor current. This refers to the angular velocity of the motor. The above parameters are obtained and stored through offline identification, correction, and storage; The coil resistance R is measured using the DC resistance method. With the motor stationary, a DC voltage of known amplitude is applied to the coil, the steady-state current is measured, and the coil resistance is calculated according to Ohm's law. The inductance L of the coil is determined by the step voltage response method. A step voltage is applied to the coil, and the response curve of the current rising with time is recorded. According to the current response formula of the first-order RL circuit, the inductance is determined by the time constant required for the current to rise to 63.2% of the steady-state value. back electromotive force coefficient Using the constant speed driving method, the rotor of the galvanometer motor is coaxially connected to the auxiliary drive motor, and the rotor is driven to rotate at a known constant angular velocity. The open circuit voltage across the motor coil is measured, and the back electromotive force coefficient is calculated. Voltage feedforward compensation Generated by superposition of three terms: ; Inductance dynamic compensation The inductor electromotive force is pre-compensated based on the rate of change of the current command in adjacent cycles to overcome the electrical inertia when the current changes rapidly. Resistance voltage drop compensation The current command from the previous cycle is used to estimate the voltage drop across the resistor corresponding to the current, thus avoiding lag in current build-up caused by resistor voltage division. Back EMF compensation The induced electromotive force generated by the coil moving in the magnetic field is pre-compensated based on the target angular velocity, so that the current loop can respond to speed changes without waiting for the current deviation to accumulate. Of the three compensations, the back EMF compensation contributes the most during high-frequency scanning, the inductance dynamic compensation contributes significantly when the current changes rapidly, and the resistor voltage drop compensation provides steady-state voltage feedforward. This embodiment employs all three parameters simultaneously to achieve the best electrical feedforward effect.

[0030] Furthermore, the acquisition of the actual position of the galvanometer and the estimation of the actual velocity and disturbance state through the state observer specifically include: The actual position information of the galvanometer is acquired by a position sensor, and the acquired position signal is filtered. The actual location information is input into the extended state observer ESO, which is a third-order linear extended state observer LESO. The extended state observer outputs three state variables: estimated position, estimated actual velocity, and total disturbance state. The estimated actual velocity is reconstructed by the extended state observer based on the system model and position measurement information, rather than being obtained by directly performing differential operations on the actual position. The total disturbance state estimates the unmodeled frictional disturbance, parameter perturbation, and external torque disturbance as a unified extended state variable. The total disturbance state is converted into an equivalent disturbance compensation current, which is used as the disturbance compensation amount to correct the feedback correction amount. In this scheme, a capacitive angle sensor detects the actual deflection angle of the galvanometer at a sampling rate of 1MHz, and the actual position is obtained after 16-point moving average filtering. ; actual location The input is a third-order linearly extended state observer (LESO), which outputs three states: To estimate the location, To estimate the actual speed, Total disturbance state: ; ; ; ; In the formula, For the observation position, To calculate the step size, This is the estimated control gain, specifically the torque constant. With moment of inertia The ratio, To control the input, the actual value is taken from the current command of the previous cycle. The actual value is determined by the observer gain using the bandwidth method; Total disturbance state The following uncertainties are estimated as a single extended state: unmodeled frictional disturbances (including static friction, Coulomb friction and nonlinear drag torque caused by the Stribeck effect), parameter perturbations (such as model deviations caused by changes in moment of inertia with lens load and the influence of temperature on damping coefficient), and external torque disturbances (such as the elastic drag torque of connecting cables and external impact vibrations). The total perturbation state output by LESO As a disturbance compensation amount, it is used to correct the feedback correction amount, so as to achieve unified compensation for multiple unknown disturbances without the need for separate modeling; The feedback correction quantity is generated using a dual-channel parallel structure; The first channel is the error feedback channel, which calculates the position deviation and then... and the actual speed estimated by LESO Error feedback components are generated by the PD controller: ; In the formula, For proportional gain, This is the differential gain; Among them, proportional gain and differential gain Tuning is performed based on the galvanometer dynamics model and the system dynamic response requirements; First, obtain the moment of inertia of the galvanometer. Damping coefficient Elasticity coefficient and torque constant The parameters mentioned above are obtained through offline system identification; Then, the closed-loop bandwidth is determined based on the mechanical resonant frequency. Damping ratio The damping ratio was optimized using an discrete step method, with its initial value set as an intermediate value within the range of 0.7 to 1.0 based on the system's dynamic response requirements. During the step response test, the damping ratio was adjusted according to whether the system oscillated and the change in convergence speed. Incremental or decremental corrections are performed, with each adjustment step ranging from 0.05 to 0.1. When the system response is too slow, each adjustment is made by changing the damping ratio. The damping ratio is gradually reduced in units. Continue until the system satisfies the condition of no sustained oscillation; Furthermore, the closed-loop characteristic equation is established and matched with the characteristic equation of a standard second-order system to calculate the proportional gain. and differential gain The initial value, where the scaling gain satisfy Differential gain satisfy ; Subsequently, the proportional gain was adjusted based on the step response test results. and differential gain Optimization and adjustments were made, including the proportional gain. Each adjustment is made by 5% to 20% of the current parameter value, and the adjustment is gradually increased proportionally when the system response is slow. When overshooting or oscillation trends occur, gradually reduce by the same proportion. Differential gain Similarly, adjustments are made in increments of 5% to 20% of the current parameter value, increasing or decreasing as needed, and increasing the value when oscillations intensify. Reduced when response is sluggish By iterating step by step until the system stabilizes and converges, the system can achieve a faster dynamic response speed and a smaller trajectory tracking error while ensuring closed-loop stability. Since this scheme uses the estimated velocity output by the extended state observer as the differential feedback quantity, it has a lower noise level compared with the traditional position differential velocity measurement method. Therefore, it can use a higher differential gain without exciting mechanical resonance, thereby improving the system damping characteristics and dynamic tracking performance. The final determined proportional gain and differential gain It is stored in the controller and automatically loaded when the system starts; Among them, the proportional gain and differential gain are tuned according to the galvanometer dynamics model and the system dynamic response requirements; The actual speed here is entirely determined by LESO's state. It is provided, rather than obtained by directly performing differential operations on the position signal; Select As a source of speed for the feedback differential term, it can improve the differential gain while avoiding the excitation of mechanical resonance; The second channel is the disturbance compensation channel, which expands the LESO state. Converted into an equivalent disturbance compensation current; The outputs of the two channels are combined using an adder to obtain the corrected feedback correction amount: ; In the formula, For error feedback components, For disturbance compensation current; This additive injection method enables the error feedback channel to maintain closed-loop stability, while the disturbance compensation channel actively cancels the total disturbance at the current moment in a feedforward manner. The two are complementary in time and superimposed in effect.

[0031] Furthermore, the galvanometer is driven to deflect according to the drive control command, and the state observer state is updated, specifically including: The torque feedforward compensation amount and the corrected feedback correction amount are weighted and fused to generate a current loop command, wherein the feedforward weight and the feedback weight are dynamically adjusted according to the operating conditions, and the weight switching adopts a linear gradual transition. The coil current is detected by a sampling resistor and a differential amplifier, and the actual current is obtained after analog-to-digital conversion. Calculate the current deviation between the current loop command and the actual current; The current deviation is adjusted in a closed loop by a PI controller to generate a corresponding voltage correction amount. The PI controller is equipped with a conditional integral anti-saturation mechanism. The voltage correction amount and the voltage feedforward compensation amount are superimposed to generate a drive control command; The drive control command is modulated by PWM to generate a drive signal, and the voltage is output to the galvanometer coil through the power circuit to drive the lens to deflect. The internal state variables of the state observer are updated using the actual position, actual current, and current loop command obtained in the current control cycle. The disturbance compensation amount for the next control cycle is generated based on the updated disturbance state; In this scheme, the current loop command is composed of torque feedforward compensation. With the corrected feedback correction amount Through weighted fusion generation, the feedback weight coefficient is increased during the closed-loop control process. The proportion of feedforward weights is reduced to enhance the system's ability to correct dynamic errors. The proportion is adjusted to avoid amplifying the feedforward prediction error, where the weight adjustment adopts a normalized constraint relationship. ; The operating conditions are determined based on the normalized results of position deviation, target angular acceleration, and current deviation. Each error is normalized to the 0-1 range before a comprehensive judgment is made. A steady-state range is defined as the comprehensive error being below 0.2. The range is set to 0.2–0.4; when the overall error is between 0.2 and 0.6, it is considered to be in the medium dynamic change range, corresponding to… The range is 0.4 to 0.7; when the target angular acceleration or overall error exceeds 0.6, it is determined to be a high-speed scanning or violent change-of-direction phase, corresponding to... Take values ​​between 0.7 and 0.9; within each interval, Continuously adjust based on the normalized error value using linear interpolation; The above adjustment process uses linear interpolation to continuously change between adjacent control cycles to avoid sudden changes in weights that could cause a jump in current command, thereby ensuring the smoothness of the drive signal and the stability of the system. Meanwhile, a precision sampling resistor, in conjunction with a differential amplifier, detects the coil current, which is then digitized by a 2MSPS analog-to-digital converter to obtain the actual current. ; The current deviation is calculated, and the current loop controller uses a PI controller to generate a voltage correction value based on the current deviation. : ; in, This is the PI proportionality coefficient. It uses PI integral coefficients and is equipped with a conditional integral anti-saturation mechanism to ensure that the controller can quickly exit the saturation region under large deviations; Based on the rotational inertia obtained from offline identification Damping coefficient Elasticity coefficient and torque constant Establish an equivalent closed-loop model and call the closed-loop bandwidth. Damping ratio ; Based on this, the initial values ​​of the controller are calculated using standard second-order system characteristic matching or pole placement methods to ensure that the closed-loop system meets the target bandwidth and damping requirements, and the PI proportional coefficient is... Integral coefficient of PI Used to achieve fast current loop tracking and steady-state error elimination; Finally, the parameters were fine-tuned and corrected through step response experiments to balance response speed and steady-state accuracy, and the optimized parameters were then solidified for online control. Voltage correction amount With voltage feedforward compensation The commands are superimposed to obtain the motor drive control commands. FPGA will The signal is converted to a 200kHz PWM signal (including symmetrical dead-time compensation), which drives the full-bridge power circuit to output the corresponding voltage to the galvanometer coil. The coil generates current under the action of voltage, and the current generates Ampere torque in the permanent magnet field, which drives the lens to deflect. After the current control cycle ends, perform a state update on the state observer: The actual position of this period and current command Substitute into the LESO state update equation and calculate ; Updated expansion status Retained for the next cycle, serving as the basis for calculating the disturbance compensation amount in the next cycle; After the update is completed, the next control cycle begins, forming a complete closed-loop control cycle that continuously tracks the target trajectory with high precision.

[0032] refer to Figure 4 As shown, a current-closed-loop-based galvanometer driving accuracy stabilization control system is used to implement the aforementioned galvanometer driving accuracy stabilization control method, characterized in that it includes: The trajectory feedforward generation module is used to acquire the target trajectory of the galvanometer and calculate the target angular velocity and target angular acceleration. It generates torque feedforward compensation based on the galvanometer dynamics model and voltage feedforward compensation based on the galvanometer motor voltage model. The observation compensation module is used to acquire the actual position of the galvanometer and estimate the actual velocity and disturbance state through the state observer, generate feedback correction amount based on the position deviation and use the disturbance state for correction; The current closed-loop drive module is used to fuse the corrected feedback correction amount with the torque feedforward compensation amount to generate a current loop command, and generate a voltage correction amount based on the current deviation and fuse it with the voltage feedforward compensation amount to generate a drive control command to drive the galvanometer to perform deflection. The state iteration module is used to update the internal state variables of the state observer and generate the disturbance compensation amount for the next control cycle, thereby realizing the closed-loop iteration of the control cycle.

[0033] Furthermore, the trajectory feedforward generation module includes a target trajectory processing unit, a torque feedforward calculation unit, and a voltage feedforward calculation unit: The target trajectory processing unit is used to receive target position trajectory data sent by the upper control system, calculate the target angular velocity and target angular acceleration through differential operation of adjacent trajectory points, perform low-pass filtering after differential operation, and use the calculation results as feedforward control input parameters. The torque feedforward calculation unit is used to obtain the moment of inertia, damping coefficient, elastic coefficient obtained through offline system identification experiments, and the torque constant measured through constant current deflection experiments, establish a galvanometer dynamic model, and calculate the inertial torque compensation term, damping torque compensation term, and elastic restoring torque compensation term respectively, and generate the torque feedforward compensation amount. The voltage feedforward calculation unit is used to obtain the resistance parameters, inductance parameters, and back electromotive force parameters of the galvanometer motor, and generate inductance dynamic compensation, resistance voltage drop compensation, and back electromotive force compensation respectively. At least one compensation is superimposed as the voltage feedforward compensation.

[0034] Furthermore, the observation compensation module includes an actual position acquisition unit, a state observation unit, and a feedback correction unit: The actual position acquisition unit is used to obtain the actual deflection angle of the galvanometer through the position sensor, and then obtain the actual position after filtering. The state observation unit is used to input the actual position into the third-order linear extended state observer, and use the extended state observer to output the estimated position, estimated actual velocity and total disturbance state. The estimated actual velocity is reconstructed by the extended state observer based on the system model and position measurement information. The total disturbance state is estimated as a unified extended state variable by combining the unmodeled friction disturbance, parameter perturbation and external torque disturbance. The feedback correction unit is used to calculate the position deviation between the target position and the actual position. The feedback controller generates an error feedback component based on the position deviation and the actual velocity estimated by the extended state observer. The total disturbance state is converted into an equivalent disturbance compensation current, and the corrected feedback correction is obtained by additive synthesis.

[0035] Furthermore, the current closed-loop drive module includes an instruction fusion unit, a current adjustment unit, and a drive output unit, and the state iteration module includes a state update unit, a disturbance iteration unit, and a cycle triggering unit. The instruction fusion unit is used to generate a current loop instruction by weighted fusion of the torque feedforward compensation amount and the corrected feedback correction amount. The feedforward weight and feedback weight are dynamically adjusted according to the operating conditions. The current regulation unit is used to detect the coil current through a sampling resistor and a differential amplifier, obtain the actual current after analog-to-digital conversion, calculate the current deviation between the current loop command and the actual current, and generate a voltage correction amount by performing closed-loop regulation of the current deviation through a PI controller. The drive output unit is used to superimpose the voltage correction amount and the voltage feedforward compensation amount to obtain the drive control command, and generate a drive signal through PWM modulation to control the power circuit to output voltage to the galvanometer coil, thereby driving the lens to deflect. The state update unit is used to update the internal state variables of the extended state observer using the actual position, actual current and current loop command obtained in the current control cycle. The disturbance iteration unit is used to retain the updated total disturbance state to the next control cycle and convert it into a disturbance compensation amount; The loop triggering unit is used to trigger the next control cycle after the state update is completed, and return to the trajectory feedforward generation module to continue execution, forming a closed loop iteration.

[0036] The advantages of this invention lie in its construction of a dual-feedforward compensation mechanism based on a galvanometer dynamics model and a motor voltage model. This mechanism pre-compensates for inertial torque, damping torque, and elastic restoring torque at the mechanical level, and pre-compensates for resistance voltage drop, inductance dynamic effects, and back electromotive force at the electrical level. This allows the galvanometer system to obtain the required driving force and voltage in advance when executing the target trajectory, reducing the adjustment burden on the feedback control loop. Simultaneously, it utilizes an extended state observer to perform real-time observation and unified estimation of unmodeled frictional disturbances, parameter perturbations, and external torque disturbances. The estimated disturbance states are then introduced into the feedback correction process, achieving proactive compensation for complex disturbance factors and improving the system's anti-disturbance capability and environmental adaptability. Furthermore, by establishing a current closed-loop control structure, the actual current of the galvanometer motor can quickly and accurately track the target current command, ensuring that the motor output torque remains consistent with the control requirements. Combined with the iterative update mechanism of the state observer, the system state estimation results are continuously corrected, forming a complete control closed loop from target trajectory generation, feedforward compensation, disturbance observation, closed-loop control to state update. This effectively solves the problems of dynamic response lag, large trajectory tracking error, insufficient anti-disturbance capability, and poor long-term operational stability in existing galvanometer control technologies, significantly improving the trajectory tracking accuracy, dynamic response performance, and operational stability of the galvanometer under high-speed scanning, high-precision positioning, and complex working conditions.

[0037] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A galvanometer driving precision stability control method based on current closed loop, characterized in that, include: Obtain the target position trajectory of the galvanometer, and calculate the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamic model; The torque feedforward compensation amount is generated based on the target position, target angular velocity and target angular acceleration, and the voltage feedforward compensation amount is generated based on the galvanometer motor voltage model; The actual position of the galvanometer is acquired, and the actual velocity and disturbance state are estimated through the state observer. The disturbance state includes unmodeled friction disturbance, parameter perturbation, and external torque disturbance. A feedback correction amount is generated based on the positional deviation between the target position and the actual position, and a disturbance compensation amount is generated using the disturbance state to correct the feedback correction amount. The corrected feedback correction amount is combined with the torque feedforward compensation amount to obtain the current loop command; The actual current of the galvanometer motor is obtained, and a voltage correction amount is generated by the current loop controller based on the current deviation between the current loop command and the actual current. The voltage correction amount and the voltage feedforward compensation amount are combined to obtain the galvanometer motor drive control command; The galvanometer is driven to deflect according to the drive control command, and the state observer state is updated to enter the next control cycle.

2. The galvanometer driving precision stability control method based on current closed loop according to claim 1, characterized in that, The process of acquiring the target position trajectory of the galvanometer and calculating the target angular velocity and target angular acceleration based on the target position trajectory and the galvanometer dynamics model specifically includes: Acquire target position trajectory data sent by the upper control system, wherein the target position trajectory is a discrete time sequence sampled at a fixed control period; Within each control cycle, the target position of the current cycle and the target position of the previous cycle are read. The target angular velocity is calculated by first-order backward difference, and the target angular acceleration is calculated by second-order difference based on the target angular velocity of the current cycle and the target angular velocity of the previous cycle. The results of the differential operation are low-pass filtered to suppress high-frequency quantization noise introduced by the differential operation; When the target trajectory is a known analytical function, the target angular velocity and target angular acceleration can be obtained directly by differentiating the analytical expression. The calculated target position, target angular velocity, and target angular acceleration are used as input parameters for the galvanometer feedforward control.

3. The galvanometer driving precision stability control method based on current closed loop according to claim 2, characterized in that, Based on the target position, target angular velocity, and target angular acceleration, a torque feedforward compensation amount is generated, specifically including: The rotational inertia, damping coefficient, and elastic coefficient of the galvanometer were obtained through an offline system identification experiment. The offline system identification experiment included applying a sweep frequency current excitation signal to the galvanometer, synchronously collecting deflection angle response data, and using the least squares method to fit the second-order transfer function for parameter identification. The torque constant of the galvanometer was determined independently using a constant current deflection experiment; A galvanometer dynamic model is established based on the target position, target angular velocity, and target angular acceleration. The galvanometer dynamic model is a second-order inertial-damped-elastic system model. Based on the galvanometer dynamics model, and in conjunction with the moment of inertia, damping coefficient, elastic coefficient, and torque constant, the inertial torque compensation term, damping torque compensation term, and elastic restoring torque compensation term are calculated respectively to obtain the torque feedforward compensation amount.

4. The method for stabilizing the accuracy of a galvanometer drive based on a current closed-loop control according to claim 3, characterized in that, The generation of voltage feedforward compensation based on the galvanometer motor voltage model specifically includes: Obtain the resistance, inductance, and back electromotive force parameters of the galvanometer motor; Based on the rate of change of current command in adjacent cycles, calculate the dynamic inductance compensation amount used to overcome electrical inertia when the current changes rapidly. Based on the current command of the previous cycle, calculate the resistor voltage drop compensation amount to avoid current build-up lag caused by resistor voltage division. Based on the target angular velocity, calculate the back electromotive force compensation amount used to pre-counteract the induced electromotive force generated by the coil motion; The voltage feedforward compensation is obtained by superimposing at least one of the inductance dynamic compensation, resistance voltage drop compensation, and back electromotive force compensation.

5. The method for stabilizing the accuracy of a galvanometer drive based on a current closed-loop control according to claim 4, characterized in that, The acquisition of the actual position of the galvanometer and the estimation of the actual velocity and disturbance state through the state observer specifically include: The actual position information of the galvanometer is acquired by a position sensor, and the acquired position signal is filtered. The actual location information is input into the extended state observer ESO, which is a third-order linear extended state observer LESO. The extended state observer outputs three state variables: estimated position, estimated actual velocity, and total disturbance state. The estimated actual velocity is reconstructed by the extended state observer based on the system model and position measurement information, rather than obtained by directly performing differential operations on the actual position. The total disturbance state estimates the unmodeled frictional disturbance, parameter perturbation, and external torque disturbance as a unified extended state variable. The total disturbance state is converted into an equivalent disturbance compensation current, which is used as the disturbance compensation amount to correct the feedback correction amount.

6. The method for stabilizing the accuracy of a galvanometer drive based on a current closed-loop control according to claim 5, characterized in that, The drive control command drives the galvanometer to deflect and updates the state observer state, specifically including: The torque feedforward compensation amount and the corrected feedback correction amount are weighted and fused to generate a current loop command, wherein the feedforward weight and the feedback weight are dynamically adjusted according to the operating conditions, and the weight switching adopts a linear gradual transition. The coil current is detected by a sampling resistor and a differential amplifier, and the actual current is obtained after analog-to-digital conversion. Calculate the current deviation between the current loop command and the actual current; The current deviation is adjusted in a closed loop by a PI controller to generate a corresponding voltage correction amount. The PI controller is equipped with a conditional integral anti-saturation mechanism. The voltage correction amount and the voltage feedforward compensation amount are superimposed to generate a drive control command; The drive control command is modulated by PWM to generate a drive signal, and the voltage is output to the galvanometer coil through the power circuit to drive the lens to deflect. The internal state variables of the state observer are updated using the actual position, actual current, and current loop command obtained in the current control cycle. The disturbance compensation amount for the next control cycle is generated based on the updated disturbance state.

7. A current-closed-loop-based galvanometer driving accuracy stabilization control system, used to implement the galvanometer driving accuracy stabilization control method as described in any one of claims 1-6, characterized in that, include: The trajectory feedforward generation module is used to acquire the target trajectory of the galvanometer and calculate the target angular velocity and target angular acceleration. It generates torque feedforward compensation based on the galvanometer dynamics model and voltage feedforward compensation based on the galvanometer motor voltage model. The observation compensation module is used to acquire the actual position of the galvanometer and estimate the actual velocity and disturbance state through the state observer, generate feedback correction amount based on the position deviation and use the disturbance state for correction; The current closed-loop drive module is used to fuse the corrected feedback correction amount with the torque feedforward compensation amount to generate a current loop command, and generate a voltage correction amount based on the current deviation and fuse it with the voltage feedforward compensation amount to generate a drive control command to drive the galvanometer to perform deflection. The state iteration module is used to update the internal state variables of the state observer and generate the disturbance compensation amount for the next control cycle, thereby realizing the closed-loop iteration of the control cycle.

8. The galvanometer driving accuracy stabilization control system based on current closed-loop as described in claim 7, characterized in that, The trajectory feedforward generation module includes a target trajectory processing unit, a torque feedforward calculation unit, and a voltage feedforward calculation unit. The target trajectory processing unit is used to receive target position trajectory data sent by the upper control system, calculate the target angular velocity and target angular acceleration through differential operation of adjacent trajectory points, perform low-pass filtering after differential operation, and use the calculation results as feedforward control input parameters. The torque feedforward calculation unit is used to obtain the moment of inertia, damping coefficient, elastic coefficient obtained through offline system identification experiments, and the torque constant measured through constant current deflection experiments, establish a galvanometer dynamic model, and calculate the inertial torque compensation term, damping torque compensation term, and elastic restoring torque compensation term respectively, and generate the torque feedforward compensation amount. The voltage feedforward calculation unit is used to obtain the resistance parameters, inductance parameters, and back electromotive force parameters of the galvanometer motor, and generate inductance dynamic compensation, resistance voltage drop compensation, and back electromotive force compensation respectively. At least one compensation is superimposed as the voltage feedforward compensation.

9. A galvanometer driving accuracy stabilization control system based on current closed-loop as described in claim 8, characterized in that, The observation compensation module includes an actual location acquisition unit, a status observation unit, and a feedback correction unit: The actual position acquisition unit is used to obtain the actual deflection angle of the galvanometer through the position sensor, and then obtain the actual position after filtering. The state observation unit is used to input the actual position into the third-order linear extended state observer, and use the extended state observer to output the estimated position, estimated actual velocity and total disturbance state. The estimated actual velocity is reconstructed by the extended state observer based on the system model and position measurement information. The total disturbance state is estimated as a unified extended state variable by combining the unmodeled friction disturbance, parameter perturbation and external torque disturbance. The feedback correction unit is used to calculate the position deviation between the target position and the actual position. The feedback controller generates an error feedback component based on the position deviation and the actual velocity estimated by the extended state observer. The total disturbance state is converted into an equivalent disturbance compensation current, and the corrected feedback correction amount is obtained by additive synthesis.

10. A galvanometer driving accuracy stabilization control system based on current closed-loop as described in claim 9, characterized in that, The current closed-loop drive module includes an instruction fusion unit, a current adjustment unit, and a drive output unit; the state iteration module includes a state update unit, a disturbance iteration unit, and a cycle triggering unit. The instruction fusion unit is used to generate a current loop instruction by weighted fusion of the torque feedforward compensation amount and the corrected feedback correction amount. The feedforward weight and feedback weight are dynamically adjusted according to the operating conditions. The current regulation unit is used to detect the coil current through a sampling resistor and a differential amplifier, obtain the actual current after analog-to-digital conversion, calculate the current deviation between the current loop command and the actual current, and generate a voltage correction amount by performing closed-loop regulation of the current deviation through a PI controller. The drive output unit is used to superimpose the voltage correction amount and the voltage feedforward compensation amount to obtain the drive control command, and generate a drive signal through PWM modulation to control the power circuit to output voltage to the galvanometer coil, thereby driving the lens to deflect. The state update unit is used to update the internal state variables of the extended state observer using the actual position, actual current and current loop command obtained in the current control cycle. The disturbance iteration unit is used to retain the updated total disturbance state to the next control cycle and convert it into a disturbance compensation amount; The loop triggering unit is used to trigger the next control cycle after the state update is completed, and return to the trajectory feedforward generation module to continue execution, forming a closed loop iteration.

Images