High-frequency linear electromagnet and control method thereof

Abstract

The invention discloses a high-frequency linear electromagnet and a control method thereof, and relates to the technical field of electromagnetic control, the electromagnet comprises permanent magnet steel, a magnetizer, a magnet yoke, a motion shaft, an armature, a control coil and an elastic piece, the magnetizer and the armature form four groups of uniform air gaps, after the coil is electrified, the air gap magnetic field changes, electromagnetic acting force is generated, and the motion shaft is driven to move; the electromagnetic acting force is balanced with the reset force and the loading force of the elastic piece; the control method comprises the steps of prepositive verification, magnetic field state sensing, dual-mode driving adjustment, dynamic error feed-forward compensation, stability verification and adaptive parameter updating, interference such as hysteresis and temperature is counteracted by intelligently switching a quick response mode and a high-linearity mode and combining error compensation and adaptive parameter updating, and the stability of the magnetic field is improved. The invention has the advantages of fast high-frequency response, high control precision and stable operation, can be widely applied to high-precision high-frequency driving scenes, and improves the processing quality and efficiency.

Description

Technical Field

[0001] This invention relates to the field of electromagnets and electromagnetic control technology, and more specifically to a high-frequency linear electromagnet and its control method. Background Technology

[0002] Electromechanical converters are widely used in various electromechanical systems. Taking linear electromagnets as an example, moving-iron force (torque) motors are often used in systems with fast response and high precision, such as the pre-stage drive of electro-hydraulic servo valves; moving-coil electromagnets are often used in devices with large strokes and low frequency response requirements, such as the focusing mechanism of imaging systems. In recent years, with the development of electromechanical systems, especially precision and ultra-precision machining systems, increasingly higher requirements have been placed on the performance of electromechanical converters. Taking microstructure machining systems as an example, the key component of a single-point diamond turning system is a rapid tool servo device, and the core component of this device is the electromechanical converter. When the system is working, the electromechanical converter drives the diamond tool to perform high-speed cutting on the workpiece, thereby forming microstructures with surface accuracy at the sub-micron level and surface roughness at the nanometer level on the workpiece surface. As the core actuator, the performance of the electromechanical converter, especially its frequency response, linearity, output force, and displacement, is one of the decisive factors affecting the overall performance of the system.

[0003] Piezoelectric ceramics and giant magnetostrictive materials possess advantages such as high operating frequency and large output force. Many micro-displacement electromechanical transducers use them as actuators and have been successfully applied in precision electromechanical systems such as microstructure fabrication. However, piezoelectric ceramics and giant magnetostrictive materials exhibit significant hysteresis. This nonlinear behavior causes a large amount of energy to be converted into heat, resulting in a substantial reduction in effective stroke, especially at high frequencies. Therefore, developing high-frequency electromechanical transducers with large displacements using piezoelectric ceramics and giant magnetostrictive materials presents challenges.

[0004] Moving-iron electromechanical converters are characterized by high frequency response, good linearity, and large stroke. Existing hybrid rotary or linear rapid tool servo devices use moving-iron electromechanical converters composed of multiple springs and multiple coils. However, the structure is relatively complex, requiring the integration of multiple permanent magnets or multiple coils, making assembly, adjustment, and control difficult. A search revealed that invention patent CN2004100664066 discloses a force motor with a compact structure and high frequency response, but its output linear range is narrow and its linearity is poor. Furthermore, it uses dual permanent magnets, resulting in a complex structure and difficulties in installation and testing. Invention patent CN2012100610060 discloses an electromechanical converter using an elastic device as a high-stiffness spring, which has a high frequency response, but the output is all angular displacement rather than linear displacement.

[0005] The above problems indicate that existing technologies still have significant shortcomings in achieving synergistic optimization of high frequency, high linearity, and low power consumption. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a high-frequency linear electromagnet and its control method. The goal is to overcome the inherent technical contradictions between high frequency, large stroke, high linearity, and strong robustness in existing moving-iron electromagnetic drive devices. Therefore, this invention proposes a high-frequency linear electromagnet with a symmetrical structure, controllable magnetic circuit, and air gap differential modulation. It also develops an intelligent control method integrating multi-source state perception, dual-mode drive switching, dynamic feedforward compensation, and parameter self-learning mechanisms, thereby resolving the conflict between high-frequency response and linear accuracy at the principle level.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A high-frequency linear electromagnet includes a permanent magnet, a magnetic conductor, a magnetic yoke, a motion shaft, an armature, a control coil, and an elastic element. The permanent magnet and the magnetic yoke are both located on one side of the magnetic conductor. A magnetic space is provided inside the magnetic conductor. The armature is located in the magnetic space and has an air gap between it and the magnetic conductor. The control coil is wound on the armature. The motion shaft passes vertically through the armature and is fixedly connected to the armature. The elastic element is located at one end of the motion shaft. When current is applied to the control coil, the air gap magnetic field between the armature and the magnetic conductor changes, generating electromagnetic force that drives the motion shaft to move. When the electromagnetic force is balanced with the restoring force generated by the deformation of the claw spring and the load force on the motion shaft, a new steady state is reached.

[0009] Furthermore, the magnetic conductor includes a left magnetic conductor and a right magnetic conductor, which are arranged opposite to each other. Each of the left and right magnetic conductors has an extension on its opposite side. The end of the armature is located between the extension of the left magnetic conductor and the extension of the right magnetic conductor, respectively. A first air gap and a second air gap are provided between the extension of the left magnetic conductor and the armature, and a third air gap and a fourth air gap are provided between the extension of the right magnetic conductor and the armature.

[0010] A control method for a high-frequency linear electromagnet includes the following steps: The magnetic field state sensing step involves real-time acquisition of control coil current, armature displacement, and temperature signals to construct a multi-dimensional state vector reflecting the current working state of the electromagnet. The dual-mode drive adjustment step determines the current operating mode of the electromagnet based on the multi-dimensional state vector. If it is in the step response or high-frequency tracking stage, the fast response mode based on current change rate enhancement is activated; if it is in the steady-state positioning or low-frequency scanning stage, it switches to the high linearity mode based on displacement and current mapping correction. In the dynamic error feedforward compensation step, in fast response mode, the rising edge timing characteristics are extracted based on historical step response data, and a feedforward compensation signal synchronized with the input command is generated and superimposed on the driving voltage; in high linearity mode, the compensation correction amount is output based on the deviation between the real-time displacement and the target displacement, combined with the nonlinear hysteresis loop model. The stability verification step continuously monitors the amplitude of drive current fluctuation and displacement tracking residual. When the product of the two exceeds the preset stability boundary, the mode switching delay mechanism is automatically triggered. The adaptive parameter update step involves updating the parameters in the hysteresis compensation model and the temporal weights of the feedforward signal based on the residual distribution between the actual displacement trajectory and the ideal trajectory during each complete motion cycle.

[0011] Furthermore, it also includes a pre-verification step, which involves performing two no-load displacement tests under the same step excitation after the electromagnet is started. By comparing the deviation of the actual displacement curves of the control coils under the same temperature and no-load conditions in the two tests, the main cause of displacement error is determined. The main cause of displacement error includes the mass deviation factor of the moving component and the temperature interference factor. An adjustment strategy is selected based on the main cause of displacement error. The adjustment strategy includes a mass self-correction strategy and a temperature and electricity compensation synergy strategy.

[0012] Furthermore, the dual-mode drive adjustment step includes: The fast response activation sub-step activates the current differential positive feedback path and increases the dynamic gain of the power amplifier when the rate of change of the input command exceeds a preset threshold and the displacement has not yet entered the steady-state range. In the high linearity locking sub-step, when the displacement change rate is lower than the set value and the current fluctuation amplitude is stable within the preset narrow band range, the current differential positive feedback path is closed, and the drive signal is corrected by nonlinear pre-distortion compensation according to the preset displacement current mapping table. The mode smooth transition sub-step involves progressively superimposing the two control signals through an exponentially weighted fusion algorithm during the switching process between the fast response mode and the high linearity mode, and achieving a smooth transition of the control signals by dynamically adjusting the weight ratio.

[0013] Furthermore, the dynamic error feedforward compensation step includes: The fast response compensation sub-step extracts timing features such as rising edge slope and peak delay from historical step response data, constructs a feedforward compensation signal model to generate the compensation signal synchronously with the input command, and then amplifies it proportionally before superimposing it onto the driving voltage. The high linearity compensation sub-step inputs the deviation between the real-time displacement and the target displacement into a preset nonlinear hysteresis loop model to calculate the error correction amount, and converts the correction amount into a voltage compensation signal to dynamically adjust the driving current. The compensation parameter adaptation sub-step adjusts the amplitude gain and phase shift of the feedforward compensation signal according to the dynamic characteristics of the current operating mode. In fast response mode, it increases the instantaneous peak value of the compensation signal, and in high linearity mode, it optimizes the smoothness of the compensation signal.

[0014] Furthermore, the quality self-calibration strategy includes: The actual mass determination steps involve determining the corresponding average displacement deviation based on the actual displacement curves under two no-load conditions, and deriving the actual mass of the motion component based on the preset theoretical displacement deviation upper limit and the preset theoretical mass of the motion component. The proportional gain correction sub-step updates the mass parameter in the electromagnet's motion equation to the actual mass and corrects the proportional gain of the current differential positive feedback based on the ratio of the actual mass to the theoretical mass.

[0015] Furthermore, the temperature and electricity compensation synergy strategy includes: The database establishment sub-step involves constructing a first mapping table of temperature, maximum allowable current, and maximum allowable voltage based on the safe operating current of the control coil at different temperatures in historical data, and constructing a second mapping table of temperature, hysteresis error, and compensation correction coefficient based on the hysteresis of the electromagnet at different temperatures in historical data. Finally, a temperature and hysteresis error model is constructed through quadratic polynomial fitting. The compensation calculation sub-step involves synchronously querying the first and second mapping tables based on the real-time temperature of the acquisition control coil to obtain the corresponding maximum allowable current, maximum allowable voltage, and compensation correction coefficient. The maximum allowable current and maximum allowable voltage are then converted into current-limiting control signals and sent to the control terminal. Finally, the compensation correction coefficient and the real-time displacement compensation are used to calculate the corrected compensation amount using a correction formula.

[0016] Furthermore, the stability verification step includes setting a weighted product threshold of current fluctuation energy and displacement residual square. When the product exceeds the weighted product threshold in multiple consecutive sampling periods, it is determined to be in a critical oscillation state, the holding time of the current driving mode is automatically extended, and immediate switching to another mode is prohibited until the product falls back to the safe range.

[0017] Furthermore, the adaptive parameter update step includes, after each complete motion cycle, calculating displacement residual values ​​at each sampling point based on the actual displacement trajectory data and the preset ideal displacement trajectory data to form a residual sequence, performing statistical analysis on the residual sequence, extracting the mean, variance, peak value, and distribution interval characteristics of the residuals, and determining whether the residuals are systematic deviations or random fluctuations. If they are systematic deviations, the hysteresis compensation model parameters are corrected based on the residual characteristics; if they are random fluctuations, the timing weights of the feedforward signal are adjusted.

[0018] The beneficial effects of this invention are as follows: 1. The high-frequency linear electromagnet in this invention, through the symmetrical arrangement of the left and right guiding magnets and the multi-air gap design, combined with the stacked armature and guiding magnet structure, effectively reduces eddy current loss and hysteresis loss, and improves high-frequency response performance. The double support structure composed of high-rigidity claw springs and ceramic sliding sleeves not only ensures stable support of the motion axis, but also greatly reduces frictional resistance. Combined with the lightweight motion axis design, it significantly reduces motion inertia, providing a structural foundation for high-frequency and high-precision motion. In addition, the polarized magnetic field generated by the neodymium iron boron permanent magnet makes the initial positioning accuracy of the armature high, and the differential superposition magnetic field working mode further improves the electromagnetic force output efficiency and linearity.

[0019] 2. The control method in this invention effectively improves the response speed of the electromagnet by using the differential positive feedback control method of the control coil current. Furthermore, the real-time displacement compensation control method compensates for the nonlinearity caused by hysteresis characteristics in real time, improving the linearity of the electromagnet. Specifically, the strategy combining pre-verification and dual-mode drive adjustment accurately locates the main cause of error through no-load testing. The intelligent switching and smooth transition between fast response mode and high linearity mode take into account the response speed and steady-state positioning in step response and high-frequency tracking scenarios, as well as the control accuracy in low-frequency scanning scenarios. Moreover, the dynamic error feedforward compensation and adaptive parameter update mechanism accurately offset the errors caused by hysteresis, temperature and other factors, continuously optimizes the control parameters, and ensures the control accuracy of long-term operation. In addition, the stability verification step effectively avoids the risk of critical oscillation, and the temperature and electricity compensation coordination and quality self-correction strategy specifically solve the performance fluctuation problem caused by temperature interference and component quality deviation. Attached Figure Description

[0020] Figure 1 This is a three-dimensional schematic diagram of the structural principle of the present invention; Figure 2 This is a three-dimensional schematic diagram of the motion component of the present invention; Figure 3 This is a three-dimensional schematic diagram of the magnetic conductor of the present invention; Figure 4 This is a schematic diagram of the magnetic circuit of the present invention; Figure 5 This is a simulation calculation diagram of the magnetic field distribution of the present invention; Figure 6 This is a flowchart of the control section in this invention; Figure 7 These are the electromagnetic force simulation calculation results of this invention; Figure 8 This is the static characteristic test curve of the present invention; Figure 9 This is the amplitude-frequency response test curve of the present invention; Figure 10 This is a schematic diagram of the controller of the present invention; Figure 11 The simulation results are for the current differential closed-loop control strategy of this invention. Figure 12 This is a test diagram showing the application results of the present invention.

[0021] Auxiliary markings: 1. Displacement sensor; 2. Housing; 3. Left guide magnet; 4. Left yoke; 5. Control coil; 6. Permanent magnet; 7. Armature; 8. Motion shaft; 9. Sliding sleeve; 10. Tool holder; 11. Claw spring; 12. Right yoke; 13. Right guide magnet; 14. Lower mounting plate; 15. Upper mounting plate; 16. Rectangular soft magnetic material sheet; 17. L-shaped soft magnetic material sheet; 18. Input control signal; 19. Signal synthesizer; 20. Signal superimposed unit; 21. Proportional element; 22. Power amplifier; 23. Differential element; 24. Signal processing unit; 25. Sampling resistor. Detailed Implementation

[0022] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Identical components are denoted by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "upper," and "lower" used in the following description refer to directions in the accompanying drawings, and the terms "bottom surface," "top surface," "inner," and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0023] like Figures 1-3 As shown, the high-frequency linear electromagnet of this invention includes a displacement sensor 1, a housing 2, a left guiding magnet 3, a left magnetic yoke 4, a control coil 5, a permanent magnet 6, an armature 7, a motion shaft 8, a sliding sleeve 9, a tool holder 10, a claw spring 11, a right magnetic yoke 12, and a right guiding magnet 13. The armature 7 is fixed on the motion shaft 8, together forming the motion component of the electromagnet. The armature 7 is made of multiple I-shaped soft magnetic material sheets stacked together. This structural design can effectively reduce eddy current loss and hysteresis loss, and improve the high-frequency response performance of the electromagnet. The motion shaft 8 is made of lightweight materials, such as aluminum or ceramic, to reduce the inertia of the motion component and ensure the response speed under high-frequency motion.

[0024] The elastic element is a claw spring 11. The spring plates of the claw spring 11 are fixed to the housing 2, with its outer ring fixedly connected to the housing 2 and its inner ring fixed to the motion shaft 8. The inner hole of the sliding sleeve 9 is connected to the motion shaft 8, and the sliding sleeve 9 is fixed to the housing 2. The claw spring 11 and the sliding sleeve 9 together provide double support for the motion shaft 8. The sliding sleeve 9 is made of ceramic material, which has good wear resistance and lubricity, and can reduce the frictional resistance when the motion shaft 8 moves. The claw spring 11 is composed of multiple spring plates, each of which is made of high-stiffness spring steel and has been heat-treated. The spring plates are distributed in a claw shape, which can provide stable elastic support for the motion shaft 8. When current is applied to the control coil 5, the air gap magnetic field between the armature 7 and the magnetic conductor changes, causing the motion shaft 8 to move and compress the claw spring 11. The claw spring 11 is used to balance the load force generated by the change in the air gap magnetic field and the reset force of the motion shaft 8. The stiffness of the claw spring 11 can be expressed by the formula: ,in For the elastic modulus, 65Mn spring steel Approximately 200 GPa, Poisson's ratio, These are the width, thickness, and length of the reed, respectively. For the number of reeds, The geometric constant is 0.9 in this invention. This structure makes the stiffness of the claw spring much higher than that of ordinary wound springs, and can quickly balance electromagnetic force and load force.

[0025] The left and right magnetic guides 3 and 13 are respectively formed by an upper mounting plate 15, a lower mounting plate 14, and multiple rectangular soft magnetic material sheets 16 and multiple L-shaped soft magnetic material sheets 17 stacked together. The stacked structure can further reduce eddy current losses. The left and right magnetic yokes 4 and 12 are made of soft magnetic materials and are used to guide the direction of the magnetic field and enhance the magnetic field strength. The left and right magnetic guides 3 and 13 are arranged opposite to each other, and each of them has an extension (i.e., multiple L-shaped soft magnetic material sheets 17 stacked together) on its opposite side. The ends of the armature 7 are located between the extensions of the left and right magnetic guides 3 and 13, respectively, and a first air gap is formed between the extension of the left magnetic guide 3 and the armature 7. Second air gap A third air gap is formed between the extension of the right guide magnet 13 and the armature 7. and the fourth air gap Each air gap has the same length in the initial state, which is 0.20 mm in this embodiment.

[0026] In this embodiment, the control coil 5 is made of high-temperature resistant enameled wire and is wound on the armature 7. It has 100 turns and the wire diameter is 0.5mm. The permanent magnet 6 is made of neodymium iron boron and has a size of 25. 25 20mm, fixed to one side of the magnetic conductor, used to generate a polarizing magnetic field, so that the armature 7 is in a neutral and relatively balanced state when no current is applied to the control coil 5, such as Figure 4 The magnetic field distribution diagram of the armature 7 in the present invention when it is in the neutral position, wherein the three arrows indicate the magnetic field lines of the polarization magnetic field generated by the permanent magnet 6, and the double arrows indicate the magnetic field lines of the control magnetic field generated by the control coil 5. Figure 5 The figure shown is a three-dimensional spatial magnetic field distribution diagram of the present invention obtained from simulation calculations, wherein... Figure 5 (a) and Figure 5 (b) The diagrams show the magnetic field distribution when the control coil current is 0A and 3A, respectively. As can be seen from the diagrams, the electromagnet's magnetic field is composed of the spatial three-dimensional polarized magnetic field generated by the permanent magnet 6 and the spatial three-dimensional magnetic field generated by the control coil current. Figure 4 Distribution of magnetic field lines Figure 1 To.

[0027] When a current of a certain polarity is input to control coil 5, a control magnetic field is generated. The control magnetic field and the polarization magnetic field are differentially superimposed, and the air gap... , The magnetic field in the air gap is enhanced. , As the magnetic field weakens, the forces on armature 7 become unbalanced. Under the influence of electromagnetic force, armature 7 overcomes the elastic force and load force of the claw spring 11, causing it to move towards the air gap. , The armature 7 moves a certain displacement in the direction of shortening until the electromagnetic force and the elastic force and load force of the claw spring 11 are balanced; when the polarity of the input current is opposite, the armature 7 moves in the opposite direction.

[0028] The tool holder 10 is mounted on the motion axis 8 for mounting machining tools. The displacement sensor 1 is fixed on the housing 2 for real-time acquisition of the displacement signal of the armature 7, providing feedback data for the control strategy.

[0029] The main dimensional parameters of the electromagnet prototype of this invention are shown in Table 1 below.

[0030] The permanent magnet 6 is made of neodymium iron boron and has a size of 25. 25 20mm; the length of armature 7 is 25mm, and the size of the magnetic pole face is 12mm. 8mm 2. Air gap , , , The length is 0.20mm; the number of high-stiffness springs of claw spring 11 is 4, the thickness is 0.6mm, the outer diameter is 20mm, and the inner diameter is 3mm; the number of turns of control coil 5 is 100, and the wire diameter of enameled wire is 0.3mm.

[0031] The control method for the high-frequency linear electromagnet in this invention, such as... Figure 6 As shown, it includes a pre-verification step, a magnetic field state sensing step, a dual-mode drive adjustment step, a dynamic error feedforward compensation step, a stability verification step, and an adaptive parameter update step.

[0032] The control method is partly based on, for example Figure 10 The schematic diagram of the controller of the present invention shown includes an input control signal 18, a signal synthesizer 19, a signal superimposed unit 20, a proportional element 21, a power amplifier 22, a differential element 23, a signal processing unit 24, and a sampling resistor 25.

[0033] In the pre-verification step, after the electromagnet is started, two no-load displacement tests are performed under the same step excitation. Both tests are conducted under conditions where the control coil temperature is stable and there is no external load. By comparing the deviation of the two actual displacement curves, the main cause of displacement error is determined. If the absolute value of the average deviation of the two displacement curves is greater than a preset threshold and the deviation direction is consistent, the main cause of error is determined to be the mass deviation factor of the moving component. If the deviation of the two displacement curves shows a regular drift with temperature changes, the main cause of error is determined to be the temperature interference factor. Based on the determination result, the corresponding adjustment strategy is selected: if it is the mass deviation factor, the mass self-correction strategy is activated; if it is the temperature interference factor, the temperature-electric compensation synergy strategy is activated. The pre-verification step can quickly locate the root cause of error in the early stage of electromagnet startup, providing a basis for the precise adjustment of subsequent control strategies and avoiding the decrease in control accuracy caused by blind adjustment.

[0034] In the quality self-calibration strategy, the corresponding average displacement deviation is calculated based on the actual displacement curves under two no-load conditions. and Take their average value According to the preset theoretical upper limit value of displacement deviation Theoretical mass of moving components Derive the actual mass of the moving components The calculation relationship is as follows: Subsequently, the mass parameter in the electromagnet's equation of motion was updated to... And based on the ratio of actual quality to theoretical quality. Corrected current derivative positive feedback proportional gain ,Right now The quality self-calibration strategy can reduce displacement errors caused by mass deviations of moving components, and the precise correction of proportional gain ensures response speed and control accuracy in fast response mode.

[0035] In the temperature-electricity compensation coordinated strategy, the system pre-constructs two mapping tables based on historical operating data: the first mapping table shows the correspondence between temperature and maximum allowable current and maximum allowable voltage; the second mapping table shows the correspondence between temperature and hysteresis error and compensation correction coefficient. Simultaneously, a temperature-electricity compensation coefficient is constructed through quadratic polynomial fitting. With hysteresis error The model: ,in, The fitting coefficients are used to collect the temperature of the control coil in real time during operation. Simultaneously query the first and second mapping tables to obtain the corresponding maximum allowable current. Maximum allowable voltage and compensation correction coefficient ,Will and Converted into a current-limiting control signal and sent to the power amplifier control terminal, With real-time displacement compensation By correcting the formula: The corrected compensation amount was calculated. For subsequent signal synthesis, the temperature-electric compensation synergy strategy can reduce the hysteresis error caused by temperature changes, and the current limiting control ensures the safe operation of the coil at different temperatures, avoiding overheating damage. At the same time, the accurate calculation of the compensation amount improves the stability and accuracy of displacement control.

[0036] In the magnetic field state sensing step, a displacement sensor is fixed to the housing, and its real-time output signal is input to the signal processing unit after A / D conversion; a sampling resistor is connected in series in the control coil circuit to collect the coil current signal in real time; a temperature sensor is attached to the outer wall of the coil to collect the coil temperature signal in real time. These three components together constitute a multi-dimensional state vector. As the basis for subsequent control decisions, the multidimensional state vector comprehensively reflects the working state of the electromagnet, providing reliable data support for rapid decision-making in control strategies. Compared with single state parameter acquisition, the accuracy of control decisions is effectively improved.

[0037] In the dual-mode drive adjustment step, the current operating mode is determined based on the multi-dimensional state vector. If the rate of change of the input command is... Exceeding the preset threshold If the displacement has not yet entered the steady-state region, then the fast response mode is activated; if the displacement change rate And the current fluctuation amplitude Then, it switches to high linearity mode. The dual-mode drive adjustment can adaptively select the control mode according to the needs of different operating stages of the electromagnet, so as to balance fast response and high linearity control, and solve the problem that the traditional single control mode cannot simultaneously meet the requirements of response speed and control accuracy.

[0038] In the fast response activation sub-step, the activation current differential positive feedback path is activated: for the sampled current Performing differentiation operations yields Then scaled up proportionally Then superimposed on the original control signal Generate enhanced signal At the same time, the dynamic gain of the power amplifier is increased to accelerate the current build-up speed. The increase in the dynamic gain of the power amplifier further accelerates the response speed and shortens the step response rise time, meeting the requirements of high-frequency tracking and step response scenarios.

[0039] In the high-linearity locking sub-step, the current differential positive feedback path is closed, and only the basic control signal is retained. The signal processing unit then processes the signal according to a preset displacement-current mapping table. For target displacement Perform nonlinear pre-distortion compensation to generate a corrected driving signal. This compensates for displacement distortion caused by magnetic circuit nonlinearity, meeting the high linearity requirements of steady-state positioning and low-frequency scanning scenarios.

[0040] In the mode smooth transition sub-step, an exponentially weighted fusion algorithm is used to progressively superimpose the two control signals. Let the weight at the current time be... The synthesized control signal is then: , in The control signal decays from 1 to 0 or from 0 to 1 according to an exponential law. The decay or growth time constant is dynamically set according to the mode switching direction to ensure that the control signal does not jump.

[0041] In the dynamic error feedforward compensation step, different feedforward compensation strategies are selected according to the current operating mode. In fast response mode, the system extracts the rising edge slope from the historical step response database. and peak latency Construct a feedforward compensation signal model based on equal time-series characteristics. ,in As a proportional gain, the compensation signal is generated synchronously with the input command and then amplified proportionally before being superimposed on the drive voltage.

[0042] In high linearity mode, real-time displacement With target displacement deviation Input a preset nonlinear hysteresis loop model Calculate the error correction amount This correction amount is then converted into a voltage compensation signal. It is used to dynamically adjust the drive current. The nonlinear hysteresis loop model can accurately fit the hysteresis characteristics, effectively offset the displacement error caused by the hysteresis phenomenon, and improve the control accuracy in high linear mode.

[0043] In the compensation parameter adaptation sub-step, the amplitude gain and phase shift of the feedforward compensation signal are adjusted according to the dynamic characteristics of the current operating mode. In fast response mode, the instantaneous peak value of the compensation signal is increased to accelerate the response; in high linearity mode, a low-pass filter is applied to the compensation signal to optimize smoothness and suppress high-frequency noise.

[0044] During the stability verification step, the amplitude of drive current fluctuations is continuously monitored. Displacement tracking residual Calculate the weighted product of current fluctuation energy and the square of displacement residual, and set a threshold for the weighted product. When continuous When the weighted product of the current fluctuation energy and the square of the displacement residual within a sampling period exceeds a weighted product threshold, the system is determined to be in a critical oscillation state. At this point, the hold time of the current driving mode is automatically extended, and immediate switching to another mode is prohibited until the weighted product of the current fluctuation energy and the square of the displacement residual continuously exceeds a threshold. When the number of cycles is below the weighted product threshold, the stability verification step can effectively identify critical oscillation states with high accuracy. The mode switching delay mechanism avoids aggravated oscillations, ensures stable system operation, and reduces the risk of equipment damage.

[0045] In the adaptive parameter update step, the system acquires the actual displacement trajectory after each complete motion cycle. With respect to the preset ideal displacement trajectory Calculate the displacement residual value for each sampling point. To form a residual sequence Statistical analysis is performed on the sequence to calculate the residual mean, variance, peak value, and patching interval. If the absolute value of the residual mean is greater than the steady-state displacement interval, it is determined to be a systematic deviation; otherwise, it is determined to be random fluctuation. If it is a systematic deviation, the parameters of the hysteresis compensation model are modified according to the residual characteristics. For example, if the residual is monotonically offset, the center offset of the hysteresis loop model is adjusted; if the residual is asymmetrically distributed, the slopes of the upper and lower branches of the model are adjusted; if it is random fluctuation, the time-series weight of the feedforward signal is adjusted, for example, by increasing the sample weight of similar working conditions in the historical step response data to improve the feedforward prediction accuracy.

[0046] Figure 7As shown in the figure, the simulation calculation curves of electromagnetic force and displacement of the electromagnet in this embodiment of the invention under different control currents are as follows. It can be seen from the figure that within the control current range of ±5A, the electromagnetic force on the armature 7 is linearly proportional to the current of the control coil 5. At the same time, within the displacement range of ±60μm, the change of electromagnetic force on the armature 7 is linear. When the control current is 5A, the force on the armature reaches more than 180N.

[0047] like Figure 8 As shown in the figure, the static characteristic test curve of the electromagnet prototype in this embodiment of the invention shows that the displacement of the armature remains linear within the range of ±5A. When the current exceeds this range, it enters the saturation state. The displacement range is -62.3-61.6μm, and the hysteresis is less than 3.8%.

[0048] like Figure 9 As shown, the amplitude-frequency response characteristic test curve of the electromagnet prototype in this embodiment of the invention has a bandwidth of 9.26 kHz; when the frequency exceeds 6.9 kHz, the amplitude decreases significantly. This is mainly because the yoke, magnetic conductor, and armature of the electromagnet prototype are not laminated, resulting in larger eddy current losses and hysteresis losses.

[0049] like Figure 11 As shown, the simulation results of this invention are based on the positive feedback control strategy of the control coil current derivative. The input voltage is a step signal of 4V, the gain of the proportional amplifier is 10, and the time constant of the current derivative is 0.01ms. Figure 11 As can be seen, by employing the control strategy of this invention, the rise time of the electromagnet's step response is reduced from 0.16 ms before using the control strategy to 0.08 ms, significantly improving the electromagnet's response speed, because the differential current signal rises rapidly in the initial stage of the voltage step signal, then falls rapidly and tends to stabilize. Specific effects are described in the figure below. Figure 12 The figure shown is a test result of the actual application of the high-frequency linear electromagnet and its control method designed in this invention.

[0050] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A high-frequency linear electromagnet, characterized in that: The device includes a permanent magnet, a magnetic conductor, a magnetic yoke, a motion shaft, an armature, a control coil, and an elastic element. The permanent magnet and the magnetic yoke are both located on one side of the magnetic conductor. The magnetic conductor has a magnetic space inside. The armature is located in the magnetic space and has an air gap between it and the magnetic conductor. The control coil is wound on the armature. The motion shaft passes vertically through the armature and is fixedly connected to the armature. The elastic element is located at one end of the motion shaft. When current is applied to the control coil, the air gap magnetic field between the armature and the magnetic conductor changes, generating electromagnetic force that drives the motion shaft to move. When the electromagnetic force is balanced with the restoring force generated by the deformation of the elastic element and the load force on the motion shaft, a new steady state is reached.

2. The high-frequency linear electromagnet according to claim 1, characterized in that: The magnetic conductor includes a left magnetic conductor and a right magnetic conductor, which are arranged opposite to each other. Each of the left and right magnetic conductors has an extension on its opposite side. The end of the armature is located between the extension of the left magnetic conductor and the extension of the right magnetic conductor, respectively. A first air gap and a second air gap are provided between the extension of the left magnetic conductor and the armature, and a third air gap and a fourth air gap are provided between the extension of the right magnetic conductor and the armature.

3. A control method for a high-frequency linear electromagnet, based on the control method provided by the high-frequency linear electromagnet according to any one of claims 1-2, characterized in that: Includes the following steps: The magnetic field state sensing step involves real-time acquisition of control coil current, armature displacement, and temperature signals to construct a multi-dimensional state vector reflecting the current working state of the electromagnet. The dual-mode drive adjustment step determines the current operating mode of the electromagnet based on the multi-dimensional state vector. If it is in the step response or high-frequency tracking stage, the fast response mode based on current change rate enhancement is activated; if it is in the steady-state positioning or low-frequency scanning stage, it switches to the high linearity mode based on displacement and current mapping correction. In the dynamic error feedforward compensation step, in fast response mode, the rising edge timing characteristics are extracted from historical step response data, and a feedforward compensation signal synchronized with the input command is generated and superimposed on the driving voltage. In high linearity mode, the compensation correction amount is output based on the deviation between the real-time displacement and the target displacement, combined with the nonlinear hysteresis loop model. The stability verification step continuously monitors the amplitude of drive current fluctuation and displacement tracking residual. When the product of the two exceeds the preset stability boundary, the mode switching delay mechanism is automatically triggered. The adaptive parameter update step involves updating the parameters in the hysteresis compensation model and the temporal weights of the feedforward signal based on the residual distribution between the actual displacement trajectory and the ideal trajectory during each complete motion cycle.

4. The control method for a high-frequency linear electromagnet according to claim 3, characterized in that: It also includes a pre-verification step, which involves performing two no-load displacement tests under the same step excitation after the electromagnet is started. By comparing the deviation of the actual displacement curves of the control coils under the same temperature and no-load conditions in the two tests, the main cause of displacement error is determined. The main cause of displacement error includes the mass deviation factor of the moving component and the temperature interference factor. An adjustment strategy is selected based on the main cause of displacement error. The adjustment strategy includes a mass self-correction strategy and a temperature and electricity compensation synergy strategy.

5. The control method for a high-frequency linear electromagnet according to claim 4, characterized in that: The dual-mode drive adjustment step includes: The fast response activation sub-step activates the current differential positive feedback path and increases the dynamic gain of the power amplifier when the rate of change of the input command exceeds a preset threshold and the displacement has not yet entered the steady-state range. In the high linearity locking sub-step, when the displacement change rate is lower than the set value and the current fluctuation amplitude is stable within the preset narrow band range, the current differential positive feedback path is closed, and the drive signal is corrected by nonlinear pre-distortion compensation according to the preset displacement current mapping table. The mode smooth transition sub-step involves progressively superimposing the two control signals through an exponentially weighted fusion algorithm during the switching process between the fast response mode and the high linearity mode, and achieving a smooth transition of the control signals by dynamically adjusting the weight ratio.

6. The control method for a high-frequency linear electromagnet according to claim 5, characterized in that: The dynamic error feedforward compensation step includes: The fast response compensation sub-step extracts timing features such as rising edge slope and peak delay from historical step response data, constructs a feedforward compensation signal model to generate the compensation signal synchronously with the input command, and then amplifies it proportionally before superimposing it onto the driving voltage. The high linearity compensation sub-step inputs the deviation between the real-time displacement and the target displacement into a preset nonlinear hysteresis loop model to calculate the error correction amount, and converts the correction amount into a voltage compensation signal to dynamically adjust the driving current. The compensation parameter adaptation sub-step adjusts the amplitude gain and phase shift of the feedforward compensation signal according to the dynamic characteristics of the current operating mode. In fast response mode, it increases the instantaneous peak value of the compensation signal, and in high linearity mode, it optimizes the smoothness of the compensation signal.

7. The control method for a high-frequency linear electromagnet according to claim 6, characterized in that: The quality self-calibration strategy includes: The actual mass determination steps involve determining the corresponding average displacement deviation based on the actual displacement curves under two no-load conditions, and deriving the actual mass of the motion component based on the preset theoretical displacement deviation upper limit and the preset theoretical mass of the motion component. The proportional gain correction sub-step updates the mass parameter in the electromagnet's motion equation to the actual mass and corrects the proportional gain of the current differential positive feedback based on the ratio of the actual mass to the theoretical mass.

8. The control method for a high-frequency linear electromagnet according to claim 6, characterized in that: Temperature and electricity compensation coordination strategies include: The database establishment sub-step involves constructing a first mapping table of temperature, maximum allowable current, and maximum allowable voltage based on the safe operating current of the control coil at different temperatures in historical data, and constructing a second mapping table of temperature, hysteresis error, and compensation correction coefficient based on the hysteresis of the electromagnet at different temperatures in historical data. Finally, a temperature and hysteresis error model is constructed through quadratic polynomial fitting. The compensation calculation sub-step involves synchronously querying the first and second mapping tables based on the real-time temperature of the acquisition control coil to obtain the corresponding maximum allowable current, maximum allowable voltage, and compensation correction coefficient. The maximum allowable current and maximum allowable voltage are then converted into current-limiting control signals and sent to the control terminal. Finally, the compensation correction coefficient and the real-time displacement compensation are used to calculate the corrected compensation amount using a correction formula.

9. The control method for a high-frequency linear electromagnet according to claim 6, characterized in that: The stability verification step includes setting a weighted product threshold of current fluctuation energy and displacement residual square. When the product exceeds the weighted product threshold in multiple consecutive sampling periods, it is determined to be in a critical oscillation state. The holding time of the current driving mode is automatically extended, and the immediate switch to another mode is prohibited until the product falls back to the safe range.

10. The control method for a high-frequency linear electromagnet according to claim 6, characterized in that: The adaptive parameter update step includes, after each complete motion cycle, calculating displacement residual values ​​at each sampling point based on the actual displacement trajectory data and the preset ideal displacement trajectory data to form a residual sequence, performing statistical analysis on the residual sequence, extracting the mean, variance, peak value and distribution interval characteristics of the residuals, and determining whether the residuals are systematic deviations or random fluctuations. If they are systematic deviations, the hysteresis compensation model parameters are corrected according to the residual characteristics; if they are random fluctuations, the timing weights of the feedforward signal are adjusted.

Images