High-frequency adaptive space vector pulse width modulation method

Abstract

The invention relates to a high-frequency adaptive space vector pulse width modulation method, which comprises the following steps of: acquiring real-time operating parameters of a motor system to obtain a motor phase current vector and a rotor position signal; carrying out load characteristic identification on the motor system based on the motor phase current vector and the rotor position signal to obtain a load characteristic parameter set; performing frequency domain deconstruction on the load characteristic parameter set to obtain a switching frequency adaptive mapping relation; non-linear optimization is carried out on the space vector division sectors based on the switching frequency adaptive mapping relation, and a corrected space vector positioning result is obtained; and performing dead-zone adaptive compensation on the switching sequence of a power switch tube in the inverter according to a corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform, thereby solving the problems that the quality of the output waveform is reduced, the harmonic distortion is increased, and the accuracy of the high-frequency pulse width modulation waveform is reduced due to the fact that the traditional SVPWM is difficult to realize accurate vector positioning under the nonlinear load and variable working conditions. And stable operation of the motor is influenced.

Description

Technical Field

[0001] The present invention relates to the technical field of motors, and particularly to a high-frequency adaptive space vector pulse width modulation method. Background Art

[0002] In a motor drive system, although the traditional pulse width modulation (PWM) technology can effectively control the operating state of the motor, with the continuous improvement of the requirements for motor performance, especially in a high-frequency operating environment, the limitations of the traditional PWM method gradually emerge. For example, when the motor operates under high-frequency conditions, the problems of switching losses and electromagnetic interference become particularly prominent, which not only affects the efficiency of the system but also may shorten the service life of power electronic devices. In addition, due to the dynamic changes in the load characteristics, how to accurately identify and adjust the control strategy in real time to adapt to different working conditions has become an urgent problem to be solved.

[0003] In the face of the above challenges, although the existing space vector pulse width modulation (SVPWM) technology provides a relatively efficient solution, there are still deficiencies in dealing with complex load characteristics. Especially under non-linear loads and changing working conditions, it is difficult for traditional SVPWM to achieve accurate vector positioning, resulting in a decline in the quality of the output waveform and an increase in harmonic distortion, thereby affecting the smooth operation and overall performance of the motor. In addition, the existing methods have limited adaptability at different frequencies and cannot fully meet the requirements of modern high-performance motor drive systems.

[0004] In response to these problems, an improved high-frequency adaptive space vector pulse width modulation method has emerged. This method aims to accurately identify and dynamically adjust the load characteristics by collecting the operating parameters of the motor system in real time and using multi-harmonic analysis technology for frequency-domain deconstruction. At the same time, through the non-linear optimization of space vectors and the design of a dead-time adaptive compensation mechanism, the quality of the inverter output waveform is improved, the switching losses and electromagnetic interference are reduced, thereby enhancing the response speed, stability and efficiency of the entire motor drive system. This method provides new ideas and technical means for solving the challenges faced in the current field of motor control. Summary of the Invention

[0005] The main object of the present invention is to provide a high-frequency adaptive space vector pulse width modulation method, which solves the technical problem that under non-linear loads and changing working conditions, it is difficult for traditional SVPWM to achieve accurate vector positioning, resulting in a decline in the quality of the output waveform and an increase in harmonic distortion, thereby affecting the smooth operation of the motor.

[0006] To achieve the above object, the present invention provides a high-frequency adaptive space vector pulse width modulation method, including the following steps: Collect the real-time operating parameters of the motor system to obtain the motor phase current vector and the rotor position signal; Based on the motor phase current vector and the rotor position signal, identify the load characteristics of the motor system to obtain a set of load characteristic parameters; Perform frequency-domain deconstruction on the set of load characteristic parameters through multi-harmonic analysis to obtain a switching frequency adaptive mapping relationship; Based on the switching frequency adaptive mapping relationship, perform non-linear optimization on the preset space vector division sectors to obtain a corrected space vector positioning result; According to the corrected space vector positioning result, perform dead-time adaptive compensation on the switching sequence of the power switching tubes in the inverter to obtain a high-frequency pulse width modulation waveform.

[0007] Furthermore, collecting the real-time operating parameters of the motor system to obtain the motor phase current vector and the rotor position signal includes: Perform differential compensation sampling on the current sampling channels of the three-phase windings of the motor in the motor system to obtain the original three-phase current sequence, and perform two-dimensional coordinate mapping on the original three-phase current sequence through orthogonal decomposition technology to obtain the motor phase current vector; Perform complex vector decomposition on the motor phase current vector to obtain instantaneous vector components, and perform coordinate transformation on the instantaneous vector components through synchronous rotation transformation to obtain the stator current space vector; Perform high-resolution sampling on the motor shaft position through a preset rotor position encoder to obtain an initial mechanical angle sequence, and perform pole pair analysis and phase calibration on the initial mechanical angle sequence to obtain the rotor position signal.

[0008] Furthermore, based on the motor phase current vector and the rotor position signal, identifying the load characteristics of the motor system to obtain a set of load characteristic parameters includes: Perform phase sequence decomposition on the motor phase current vector to obtain an instantaneous current distribution sequence, and perform power characteristic analysis on the instantaneous current distribution sequence through cross-spectrum analysis to obtain a load power spectral density matrix; Calculate the motor load torque of the motor system based on the load power spectral density matrix to obtain an instantaneous torque ripple sequence, and perform spectrum reconstruction on the instantaneous torque ripple sequence through harmonic decomposition to obtain load dynamic characteristic parameters; Extract frequency-domain characteristics of the load dynamic characteristic parameters to obtain a load frequency response curve, and perform parameter fitting on the load frequency response curve through non-linear mapping to obtain a set of load characteristic parameters.

[0009] Furthermore, performing frequency-domain deconstruction on the set of load characteristic parameters through multi-harmonic analysis to obtain a switching frequency adaptive mapping relationship includes: Perform wavelet packet decomposition on the load characteristic parameter set to obtain a multi-scale frequency component sequence, and perform instantaneous phase extraction on the multi-scale frequency component sequence through Hilbert transform to obtain a harmonic feature vector group; Based on the harmonic feature vector group, perform harmonic energy distribution analysis on the motor system to obtain a frequency response characteristic matrix, and perform eigen-space mapping on the frequency response characteristic matrix through singular value decomposition technology to obtain a frequency-domain characteristic distribution map; Perform hyperbolic slice analysis on the frequency-domain characteristic distribution map to obtain a frequency modulation curve family, and perform non-linear transformation on the frequency modulation curve family through topological mapping technology to obtain a switching frequency mapping function; Based on the switching frequency mapping function, perform dynamic frequency characteristic analysis on the motor system to obtain a frequency response compensation sequence, and perform optimized reconstruction on the frequency response compensation sequence through adaptive filtering to obtain a switching frequency adaptive mapping relationship.

[0010] Further, the non-linear optimization of the preset space vector division sector based on the switching frequency adaptive mapping relationship to obtain a corrected space vector positioning result includes: Perform topological structure analysis on the switching frequency adaptive mapping relationship to obtain a sector boundary feature sequence, and perform phase compensation on the sector boundary feature sequence through complex domain transformation to obtain a sector dynamic boundary matrix; Based on the sector dynamic boundary matrix, perform non-uniform division on the preset space vector division sector to obtain a sector optimization distribution map, and perform boundary correction on the sector optimization distribution map through Lagrangian interpolation to obtain a sector correction parameter set; Perform vector space reconstruction on the sector correction parameter set to obtain a vector positioning feature group, and perform dynamic balance verification on the vector positioning feature group through Lyapunov stability analysis to obtain a vector positioning compensation sequence; Based on the vector positioning compensation sequence, perform non-linear mapping on the space vector to obtain a vector optimized positioning result, and perform phase synchronization on the vector optimized positioning result through hyperbolic function transformation to obtain a corrected space vector positioning result.

[0011] Further, the dead-time adaptive compensation for the power switch tubes in the inverter according to the corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform includes: Based on the corrected space vector positioning result, calculate the reference conduction time for the power switch tubes in the inverter to obtain a reference switching timing, and perform dead-time pre-compensation on the reference switching timing to obtain a pre-compensated switching timing, where the pre-compensated switching timing includes the conduction and turn-off times of each power switch tube; Predict the current ripple of the pre-compensated switching timing to obtain a current ripple prediction value, and non-linearly adjust the pre-compensated switching timing based on the current ripple prediction value to obtain an adjusted switching timing; Perform voltage vector error compensation on the adjusted switching timing to obtain a compensated switching timing, and generate a control signal for the inverter power switch based on the compensated switching timing, where the control signal includes drive pulses for each power switch; Apply the control signal to the inverter power switch to obtain a high-frequency pulse width modulation waveform.

[0012] Further, the calculating the reference conduction time of the inverter power switch based on the corrected space vector positioning result to obtain a reference switching timing includes: Determine a target voltage vector based on the corrected space vector positioning result, and perform coordinate transformation on the target voltage vector to obtain a voltage vector in the two-phase stationary coordinate system; Judge the sector of the voltage vector in the two-phase stationary coordinate system to obtain the sector where the target voltage vector is located, and select a corresponding switching state vector based on the sector where the target voltage vector is located; Calculate the action time of each switching state vector, obtain the action time of each switching state vector, and calculate the conduction time of the power switch based on the action time of each switching state vector and a preset switching frequency; Sort the conduction times of the power switches to obtain an initial reference switching timing, and calculate the duty cycle of each power switch for the initial reference switching timing; Generate a reference drive waveform for the power switch based on the duty cycle, and modulate the reference drive waveform to obtain a reference switching timing.

[0013] The present invention also provides a high-frequency adaptive space vector pulse width modulation system, including: An acquisition module for acquiring real-time operation parameters of the motor system to obtain a motor phase current vector and a rotor position signal; An identification module for identifying the load characteristics of the motor system based on the motor phase current vector and the rotor position signal to obtain a set of load characteristic parameters; A deconstruction module for performing frequency-domain deconstruction on the set of load characteristic parameters through multi-harmonic analysis to obtain a switching frequency adaptive mapping relationship; An optimization module for non-linearly optimizing a preset space vector division sector based on the switching frequency adaptive mapping relationship to obtain a corrected space vector positioning result; A compensation module, configured to perform dead-time adaptive compensation on the switching sequence of power switching tubes in an inverter according to the corrected space vector positioning result, so as to obtain a high-frequency pulse width modulation waveform.

[0014] The present invention also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0015] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0016] A high-frequency adaptive space vector pulse width modulation method provided by the present invention includes the following steps: collecting real-time operation parameters of a motor system to obtain a motor phase current vector and a rotor position signal; identifying the load characteristics of the motor system based on the motor phase current vector and the rotor position signal to obtain a load characteristic parameter set; performing frequency-domain deconstruction on the load characteristic parameter set to obtain a switching frequency adaptive mapping relationship; performing non-linear optimization on the space vector division sectors based on the switching frequency adaptive mapping relationship to obtain a corrected space vector positioning result; performing dead-time adaptive compensation on the switching sequence of power switching tubes in an inverter according to the corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform, which solves the technical problem that under non-linear loads and changing working conditions, it is difficult for traditional SVPWM to achieve accurate vector positioning, resulting in a decrease in the quality of the output waveform and an increase in harmonic distortion, thereby affecting the stable operation of the motor. The method realizes dead-time adaptive compensation for the switching sequence of power switching tubes in an inverter according to the corrected space vector positioning result, and generates a high-frequency pulse width modulation waveform. This mechanism can effectively reduce waveform distortion caused by the dead-time effect, reduce switching losses and electromagnetic interference, and further improve the performance of the system. Description of the Drawings

[0017] Figure 1 is a schematic diagram of the steps of the high-frequency adaptive space vector pulse width modulation method in an embodiment of the present invention; Figure 2 is a structural block diagram of the high-frequency adaptive space vector pulse width modulation device in an embodiment of the present invention; Figure 3 is a schematic structural block diagram of a computer device in an embodiment of the present invention.

[0018] The implementation, functional features, and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed Embodiments

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0020] As Figure 1 shown, Figure 1 is a schematic diagram of the steps of a high-frequency adaptive space vector pulse width modulation method in an embodiment of the present invention; An embodiment of the present invention provides a high-frequency adaptive space vector pulse width modulation method, including the following steps: Step S1, collect the real-time operating parameters of the motor system to obtain the motor phase current vector and the rotor position signal.

[0021] Specifically, in the high-frequency adaptive space vector pulse width modulation method, collecting the real-time operating parameters of the motor system to obtain the motor phase current vector and the rotor position signal is the basic step of the entire control process, and its purpose is to provide accurate data support for subsequent load characteristic identification and optimization processing. Specifically, this process is realized by sensors and data acquisition devices installed in the motor system. Among them, the acquisition of the motor phase current vector usually depends on high-precision current sensors, which can real-time monitor the current changes in the three-phase windings of the motor and convert them into electrical signals for analysis; while the acquisition of the rotor position signal requires the help of position sensors (such as encoders or resolvers), which can accurately capture the angular position and motion state of the rotor. Since the operating state of the motor will be dynamically adjusted due to factors such as load changes and speed fluctuations, the collected motor phase current vector and rotor position signal must have a high sampling rate and high resolution to ensure the real-time and accuracy of the data. For example, in an industrial automation scenario, assume that a robotic arm driven by a servo motor is performing a high-speed handling task. At this time, the working state of the motor may fluctuate rapidly due to load changes or external interference. To ensure the accuracy and stability of the robotic arm's movement, the control system needs to collect the motor phase current vector and rotor position signal in real time. The three-phase current detected by the current sensor can obtain the motor phase current vector through coordinate transformation, which can intuitively reflect the distribution of electromagnetic force inside the motor; at the same time, the rotor position signal provided by the position sensor is used to determine the instantaneous position and rotational speed information of the rotor. These data together form the basis for subsequent load characteristic identification. If there are noise or delay problems in the collected signals, it may lead to inaccurate load characteristic identification, thereby affecting the performance of the entire control system. Therefore, the design of this acquisition step must fully consider the anti-interference ability and transmission efficiency of the signals to ensure that the obtained motor phase current vector and rotor position signal can truly reflect the actual operating state of the motor.

[0022] Step S2: Based on the motor phase current vector and the rotor position signal, identify the load characteristics of the motor system to obtain a set of load characteristic parameters.

[0023] Specifically, based on the motor phase current vector and the rotor position signal, identifying the load characteristics of the motor system to obtain a set of load characteristic parameters is one of the key steps to implement the high-frequency adaptive space vector pulse width modulation method. Its core lies in analyzing the real-time collected motor phase current vector and rotor position signal to extract characteristic parameters that can reflect the current load state of the motor. Specifically, this process first requires synchronizing the collected motor phase current vector with the rotor position signal to ensure their temporal consistency, and then through mathematical modeling and algorithm analysis, converting these signals into a set of physical quantities describing the load characteristics. For example, the change in the motor phase current vector can reflect the fluctuation of the electromagnetic torque inside the motor, while the rotor position signal can be used to calculate the motor speed and acceleration. Combining the information of the two can further deduce key parameters such as the inertia of the load, the friction coefficient, and the external disturbing force. Taking a servo motor driving a robotic arm in an industrial automation scenario as an example, assume that the robotic arm is performing a high-speed handling task. When the robotic arm grabs objects of different weights, the load borne by the motor will change significantly. At this time, the control system can analyze the dynamic characteristics of the load through the real-time collected motor phase current vector and rotor position signal. For example, when the robotic arm grabs a heavier object, the amplitude of the motor phase current vector will increase significantly, and at the same time, the rotor position signal may show fluctuations or deceleration trends in speed. These information can be further processed through specific load characteristic identification algorithms (such as the state estimation algorithm based on Kalman filtering or the nonlinear fitting method based on neural networks) to obtain a set of load characteristic parameters including load inertia, damping coefficient, and external disturbance. This set of parameters can not only accurately describe the current load state but also provide an important basis for subsequent multiple harmonic analysis and the establishment of the switching frequency adaptive mapping relationship. If the load characteristic identification is inaccurate, it may lead to deviations in subsequent control strategies, thereby affecting the performance of the entire system. Therefore, the design of this step must fully consider the accuracy of signal processing and the robustness of the algorithm to ensure the authenticity and reliability of the set of load characteristic parameters.

[0024] Step S3: Through multiple harmonic analysis, perform frequency-domain deconstruction on the set of load characteristic parameters to obtain the switching frequency adaptive mapping relationship.

[0025] Specifically, performing frequency-domain deconstruction on the load characteristic parameter set through multi-harmonic analysis to obtain the switching frequency adaptive mapping relationship is an important step in implementing the high-frequency adaptive space vector pulse width modulation method. This process aims to deeply analyze the complex dynamic behavior during motor operation and optimize the control strategy accordingly. Specifically, this step first requires converting the load characteristic parameter set extracted from the motor phase current vector and rotor position signal into the frequency domain for analysis, and using the fast Fourier transform (FFT) or other advanced spectrum analysis techniques to identify various frequency components contained therein and their corresponding amplitude and phase information. These frequency components reflect various harmonic components generated by the motor under different load conditions, including but not limited to the fundamental wave and harmonic distortion parts. By carefully analyzing these frequency components, the influence of various factors such as electromagnetic torque fluctuations, mechanical vibrations, and thermal effects inside the motor system on the overall performance can be further understood. For example, in an industrial automation scenario where a servo motor drives a robotic arm to perform a high-speed handling task, due to load changes, the motor may generate complex current waveforms that contain rich harmonic information. Multi-harmonic analysis can accurately reveal the existence and intensity change laws of these harmonic components, thereby providing a basis for adjusting the switching frequency of the inverter. Based on the analysis results, a switching frequency adaptive mapping relationship can be established, that is, the operating frequency of the inverter is dynamically adjusted according to the real-time changes in the load characteristics. The advantage of this is that it can not only effectively reduce electromagnetic interference and losses caused by fixed-frequency operation, but also improve the response speed and stability of the system. For example, when the robotic arm grasps a heavier object and the motor load increases, multi-harmonic analysis will show a higher amplitude of low-frequency harmonic components, and at this time the system should automatically adjust to a lower switching frequency to reduce energy loss; on the contrary, a higher switching frequency can be used under light load conditions to improve efficiency. Therefore, through this series of steps, that is, from the acquisition of the load characteristic parameter set to multi-harmonic analysis, and finally to the determination of the switching frequency adaptive mapping relationship, refined management of the motor system is achieved, greatly enhancing its adaptability and flexibility.

[0026] Step S4, based on the switching frequency adaptive mapping relationship, perform non-linear optimization on the preset space vector division sectors to obtain the corrected space vector positioning result.

[0027] Specifically, nonlinearly optimizing the preset space vector partition sectors based on the switching frequency adaptive mapping relationship to obtain the corrected space vector positioning result is a crucial step in implementing the high-frequency adaptive space vector pulse width modulation method. The core lies in dynamically adjusting the partition and positioning of space vectors to better adapt to the real-time operating state of the motor system. Specifically, this process first needs to combine the switching frequency adaptive mapping relationship obtained through previous multiple harmonic analyses, analyze the optimal operating frequency range of the inverter under the current load characteristics, and use this information to guide the optimization of the space vector partition sectors. Traditional space vector pulse width modulation methods usually adopt a fixed sector partition method, which may not meet the requirements of high-performance control when facing complex load changes; by introducing nonlinear optimization algorithms (such as genetic algorithms, particle swarm optimization, or gradient descent methods), the positions and shapes of the sector boundaries can be dynamically adjusted according to the switching frequency adaptive mapping relationship, so that the vector distribution in each sector is more reasonable and can more accurately reflect the actual load requirements. For example, in an industrial automation scenario where a servo motor drives a robotic arm to perform a high-speed handling task, when the robotic arm grasps objects of different weights, the change in the motor load will cause significant fluctuations in the harmonic components and switching frequency requirements. At this time, nonlinearly optimizing the preset space vector partition sectors based on the switching frequency adaptive mapping relationship can effectively improve the response ability of the system. For example, when the robotic arm grasps a heavier object, the system will detect an increase in low-frequency harmonic components. At this time, the sector partition is readjusted through the optimization algorithm, making the space vector distribution denser in the low-frequency band, thereby improving the control accuracy; in the case of light load, the optimized sector partition will tend to the high-frequency band to reduce switching losses and improve efficiency. Through this series of nonlinear optimization operations, the corrected space vector positioning results are finally obtained. These results can not only more accurately reflect the actual operating state of the motor but also provide a reliable basis for subsequent dead-time compensation and high-frequency pulse width modulation waveform generation. Therefore, this process plays a connecting role in the entire control strategy. It not only makes full use of the load characteristics and switching frequency information extracted in the previous steps but also lays a solid foundation for subsequent inverter control.

[0028] Step S5, perform dead-time adaptive compensation on the switching sequence of the power switching tubes in the inverter according to the corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform.

[0029] Specifically, performing dead-time adaptive compensation on the switching sequence of the power switching tubes in the inverter according to the corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform is the last step in implementing the high-frequency adaptive space vector pulse width modulation method and is also a key step in ensuring the quality of the system output waveform and operating efficiency. Specifically, this process first requires re-planning the switching sequence and conduction time of the power switching tubes in the inverter based on the corrected space vector positioning result obtained through non-linear optimization before. Since there are inevitably switching delays and dead-time effects in the actual operation of the power switching tubes, these factors will cause distortion of the output waveform and affect the smoothness and efficiency of the motor operation. Therefore, it is necessary to dynamically adjust the conduction and cutoff moments of the switching tubes through a dead-time adaptive compensation mechanism to eliminate or reduce the impact of the dead-time effect on the waveform. For example, when the corrected space vector positioning result shows that the vector distribution in a certain sector is denser, it indicates that higher control accuracy is required in this area. At this time, the dead-time compensation algorithm will correspondingly adjust the switching sequence to make the switching of the power switching tubes smoother and more accurate. Taking the servo motor driving a robotic arm in an industrial automation scenario as an example, assume that the robotic arm is performing a high-speed handling task. When it grabs objects of different weights, the change in the motor load will cause fluctuations in the harmonic components and the required switching frequency, and these changes will ultimately affect the quality of the inverter output waveform. If the dead-time effect is not effectively compensated, it may cause the robotic arm to move unevenly, or even jitter or have a position deviation. In this case, based on the corrected space vector positioning result, the dead-time adaptive compensation mechanism will dynamically adjust the switching sequence of the power switching tubes according to the current load characteristics. For example, when the robotic arm grabs a heavier object and the system detects an increase in the low-frequency harmonic components, the dead-time compensation algorithm will finely adjust the switching sequence in the low-frequency band to reduce the waveform distortion caused by the dead-time effect; in the case of light load, more attention is paid to the compensation in the high-frequency band to reduce the switching loss and improve the overall efficiency. Through this dynamic adjustment, the finally generated high-frequency pulse width modulation waveform can not only meet the operating requirements of the motor under different working conditions, but also significantly improve the stability and response speed of the system. Therefore, this process plays a crucial role in the entire control strategy, making full use of the optimization results in the previous steps and providing guarantee for the efficient and smooth operation of the motor system.

[0030] In a specific embodiment, collecting the real-time operating parameters of the motor system to obtain the motor phase current vector and the rotor position signal includes: Performing differential compensation sampling on the current sampling channels of the three-phase windings of the motor in the motor system to obtain the original three-phase current sequence, and performing two-dimensional coordinate mapping on the original three-phase current sequence through orthogonal decomposition technology to obtain the motor phase current vector; Perform complex vector decomposition on the motor phase current vector to obtain instantaneous vector components, and perform coordinate transformation on the instantaneous vector components through synchronous rotation transformation to obtain the stator current space vector; Perform high-resolution sampling on the motor shaft position through a preset rotor position encoder to obtain an initial mechanical angle sequence, and perform pole pair analysis and phase calibration on the initial mechanical angle sequence to obtain a rotor position signal.

[0031] Specifically, in the high-frequency adaptive space vector pulse width modulation method, the process of collecting the real-time operating parameters of the motor system to obtain the motor phase current vector and the rotor position signal is the basis of the entire control strategy, aiming to ensure that the acquired data can accurately reflect the current operating state of the motor. First, in order to obtain the motor phase current vector, it is necessary to perform differential compensation sampling on the current sampling channels of the three-phase windings of the motor in the motor system. This process is achieved through high-precision current sensors, aiming to eliminate the influence of common-mode noise and other interference factors, thereby ensuring the accuracy of the original three-phase current sequence. Then, using orthogonal decomposition technology, these original three-phase current sequences are mapped into a two-dimensional coordinate system to form the motor phase current vector. This transformation not only simplifies the subsequent processing flow but also makes the current information more intuitively reflect the distribution of the electromagnetic torque inside the motor. Further, the obtained motor phase current vector is subjected to complex vector decomposition to extract the instantaneous vector components, which provides the basic data for subsequent analysis. Subsequently, the instantaneous vector components are transformed from the stationary coordinate system to the rotating coordinate system through synchronous rotation transformation to obtain the stator current space vector. This transformation is necessary because it can more accurately describe the behavior of the motor under different operating conditions. The stator current space vector includes the d-axis magnetic flux component and the q-axis torque component, and these two components respectively correspond to the establishment of the motor magnetic field and the generation of torque, which are crucial for optimizing the motor performance. For example, in an industrial automation scenario where a servo motor drives a robotic arm to perform a high-speed handling task, when the robotic arm grabs a heavy object, the load on the motor increases. At this time, the d-axis magnetic flux component and the q-axis torque component obtained through synchronous rotation transformation can help the control system more accurately adjust the magnetic field strength and output torque to ensure the smoothness and accuracy of the robotic arm movement. On the other hand, the process of obtaining the rotor position signal is also complex and crucial. By performing high-resolution sampling on the motor shaft position through a preset rotor position encoder, an initial mechanical angle sequence can be obtained. In this process, the selection of the encoder is particularly critical because it directly affects the accuracy of position measurement. For example, in the aforementioned application scenario, if a high-resolution optical encoder is used, very accurate angle information can be obtained, which is of great significance for improving the response speed and positioning accuracy of the entire system. However, the initial mechanical angle sequence alone is not sufficient to comprehensively describe the position state of the rotor. It is also necessary to perform pole pair resolution and phase calibration on it to obtain the final rotor position signal. This step not only involves physical-level angle calculations but also includes considerations of electrical characteristics, such as pole position identification, rotational speed feedback, and angle compensation coefficients. In particular, the application of the angle compensation coefficient can effectively correct the deviation caused by installation errors or environmental changes, ensuring the authenticity and reliability of the rotor position signal.For example, in the application scenario of a servo motor driving a robotic arm, assume that the robotic arm is performing a series of precise operations, such as assembling tiny parts. At this time, precise control of the motor rotor position is particularly important. Since each operation step requires a high degree of coordination and precision, it is necessary to rely on accurate rotor position signals to guide the operation of the motor. Specifically, when the robotic arm moves to a specified position to grasp a part, the control system sends instructions to the motor according to the pre-set path planning, and the motor adjusts its operating state based on the real-time updated rotor position signal. If there is any error in the position signal during this process, it may cause the robotic arm to fail to reach the target position accurately, thereby affecting the quality and efficiency of the entire operation. Therefore, through differential compensation sampling of the three-phase winding current of the motor and high-resolution sampling of the rotor position, combined with complex mathematical transformation and calibration techniques, not only can high-quality motor phase current vectors and rotor position signals be obtained, but also the overall performance of the motor drive system can be significantly improved, enabling it to operate stably and efficiently under various working conditions. In short, the design and implementation of this series of steps provide solid data support for subsequent space vector pulse width modulation, ensuring the reliability and flexibility of the entire control system.

[0032] In a specific embodiment, the load characteristic identification of the motor system is performed based on the motor phase current vector and the rotor position signal to obtain a load characteristic parameter set, including: Perform phase sequence decomposition on the motor phase current vector to obtain an instantaneous current distribution sequence, and perform power characteristic analysis on the instantaneous current distribution sequence through cross-spectrum analysis to obtain a load power spectral density matrix; Calculate the motor load torque of the motor system based on the load power spectral density matrix to obtain an instantaneous torque ripple sequence, and perform spectral reconstruction on the instantaneous torque ripple sequence through harmonic decomposition to obtain load dynamic characteristic parameters; Extract frequency-domain features from the load dynamic characteristic parameters to obtain a load frequency response curve, and perform parameter fitting on the load frequency response curve through non-linear mapping to obtain a load characteristic parameter set.

[0033] Specifically, in the high-frequency adaptive space vector width modulation method, the process of identifying the load characteristics of the motor system based on the motor phase current vector and the rotor position signal to obtain the load characteristic parameter set is a key step in achieving precise control. First, by decomposing the motor phase current vector into phase sequences, an instantaneous current distribution sequence can be obtained. This process can not only separate the fundamental component and harmonic component but also provide basic data for subsequent analysis. Specifically, by performing power characteristic analysis on the instantaneous current distribution sequence through cross-spectrum analysis, a load power spectral density matrix can be obtained. This matrix contains rich information, including but not limited to the fundamental power component, harmonic power component, and cross-frequency response coefficient. These information are of great significance for understanding the power loss mechanism during motor operation. For example, in an industrial automation scenario where a servo motor drives a robotic arm to perform a high-speed handling task, when the robotic arm grasps objects of different weights, the load on the motor will change significantly, resulting in increased fluctuations in its internal electromagnetic torque. At this time, the load power spectral density matrix obtained through cross-spectrum analysis can help engineers accurately evaluate the energy conversion efficiency of the motor under different load conditions. Further, based on the load power spectral density matrix, the motor load torque of the motor system is calculated to obtain an instantaneous torque ripple sequence. This step involves complex mathematical modeling and algorithm processing, aiming to convert the information in the load power spectral density matrix into physical quantities that can directly reflect the output characteristics of the motor. Next, by performing harmonic decomposition on the instantaneous torque ripple sequence for spectral reconstruction, load dynamic characteristic parameters can be obtained. These parameters cover important indicators such as torque volatility, load inertia coefficient, and damping ratio eigenvalue, which jointly describe the dynamic behavior of the motor under actual working conditions. For example, in the aforementioned application scenario, when the robotic arm moves quickly and grasps a heavy object, the torque volatility of the motor will increase, and at the same time, due to the change in load mass, the load inertia coefficient will also be adjusted accordingly. In this case, accurately obtaining these dynamic characteristic parameters is crucial for optimizing the performance of the control system. Subsequently, frequency domain feature extraction is performed on the load dynamic characteristic parameters to obtain a load frequency response curve. This process uses advanced spectral analysis techniques, such as Fourier transform or wavelet transform, to convert the dynamic characteristic parameters in the time domain into information in the frequency domain, making the influence of different frequency components more intuitive. Based on this load frequency response curve, parameter fitting is performed on it through non-linear mapping, and finally, a load characteristic parameter set is obtained. This parameter set not only contains the load type identifier but also involves key information such as dynamic response coefficients and steady-state characteristic indicators. For example, in the application case of a servo motor driving a robotic arm, if the robotic arm needs to operate under different load conditions frequently, then according to the real-time updated load characteristic parameter set, the control system can quickly adjust its own control strategy to ensure that each action can achieve the best effect.Especially for those tasks with extremely high precision requirements, such as precision assembly operations, any minor deviation may lead to the failure of the entire production process. Therefore, accurately identifying and applying load characteristic parameters is particularly important. In summary, through the above series of steps, starting from the motor phase current vector and rotor position signal, after a series of complex data processing processes such as phase sequence decomposition, cross-spectrum analysis, and harmonic decomposition, a parameter set that comprehensively describes the load characteristics of the motor system is finally obtained. This not only provides an important reference basis for subsequent space vector pulse width modulation but also lays a solid foundation for improving the stability and efficiency of the entire motor drive system. Especially when facing a complex and changing working environment, such as the various challenges encountered by a servo motor-driven robotic arm during the execution of diverse tasks, the load characteristic parameter set obtained by this method can help the control system respond more intelligently to various situations and ensure that each operation can be completed efficiently and accurately. Therefore, this series of technical means is not only an innovative breakthrough in theory but also an indispensable important link in practical engineering applications.

[0034] In a specific embodiment, the frequency-domain deconstruction of the load characteristic parameter set through multiple harmonic analyses to obtain a switching frequency adaptive mapping relationship includes: Performing wavelet packet decomposition on the load characteristic parameter set to obtain a multi-scale frequency component sequence, and performing instantaneous phase extraction on the multi-scale frequency component sequence through Hilbert transform to obtain a harmonic feature vector group; Based on the harmonic feature vector group, performing harmonic energy distribution analysis on the motor system to obtain a frequency response characteristic matrix, and performing eigen-space mapping on the frequency response characteristic matrix through singular value decomposition technology to obtain a frequency-domain characteristic distribution map; Performing hyperbolic slice analysis on the frequency-domain characteristic distribution map to obtain a frequency modulation curve family, and performing non-linear transformation on the frequency modulation curve family through topological mapping technology to obtain a switching frequency mapping function; Based on the switching frequency mapping function, performing dynamic frequency characteristic analysis on the motor system to obtain a frequency response compensation sequence, and performing optimized reconstruction on the frequency response compensation sequence through adaptive filtering to obtain a switching frequency adaptive mapping relationship.

[0035] Specifically, in the high-frequency adaptive space vector pulse width modulation method, the process of performing frequency-domain deconstruction on the load characteristic parameter set through multi-harmonic analysis to obtain the switching frequency adaptive mapping relationship is a key step in achieving efficient motor control. First, this process begins with wavelet packet decomposition of the load characteristic parameter set, thereby obtaining a multi-scale frequency component sequence. As a powerful time-frequency analysis tool, wavelet packet decomposition can decompose a signal into multiple components in different frequency bands, each corresponding to a specific frequency range and time resolution. For a servo motor driving a robotic arm to perform high-speed handling tasks, when the robotic arm grasps objects of different weights, the load change on the motor will cause the current signal to contain complex frequency components. Using wavelet packet decomposition can effectively separate these components to form a multi-scale frequency component sequence, providing basic data for further analysis. Next, instantaneous phase extraction is performed on the multi-scale frequency component sequence through Hilbert transform to obtain a harmonic feature vector group. Hilbert transform is a technique used for analytical signal processing that can extract the instantaneous phase information of a signal without changing the original signal amplitude. In the aforementioned application scenario, due to load changes, the electromagnetic torque fluctuations inside the motor will generate a series of harmonic components, which not only contain fundamental wave information but also rich harmonic information. The instantaneous phase information extracted through Hilbert transform can help identify these harmonic components and convert them into a harmonic feature vector group, laying a foundation for subsequent frequency response analysis. Based on the harmonic feature vector group, harmonic energy distribution analysis is performed on the motor system to obtain a frequency response characteristic matrix. This step involves a comprehensive assessment of the energy distribution of the motor system across the entire operating frequency band. Specifically, in the application scenario of a servo motor driving a robotic arm, as the load of the robotic arm changes, the energy distribution of the motor system will also change. By analyzing the harmonic feature vector group, a matrix that comprehensively describes the frequency response characteristics of the motor system can be obtained. Subsequently, singular value decomposition technology is used to perform eigen-space mapping on this frequency response characteristic matrix to obtain a frequency-domain feature distribution map. As an effective dimensionality reduction technology, singular value decomposition can convert complex high-dimensional data into an easily understandable and processed form, making the frequency response characteristics of the motor system more intuitively presented. Then, hyperbolic slice analysis is performed on the frequency-domain feature distribution map to obtain a family of frequency modulation curves. Hyperbolic slice analysis is a technique specifically used for complex signal analysis that can observe the performance of a signal in different dimensions by selecting different slice angles. In this process, the family of frequency modulation curves reflects the frequency response characteristics of the motor system under different load conditions. To further optimize these characteristics, a non-linear transformation is performed on the family of frequency modulation curves through topological mapping technology to obtain a switching frequency mapping function. This non-linear transformation aims to dynamically adjust the operating frequency of the inverter according to the actual working conditions to ensure that the motor system always operates in an optimal state.Finally, based on the switching frequency mapping function, the dynamic frequency characteristics of the motor system are analyzed to obtain a frequency response compensation sequence. This step involves a detailed analysis of the response characteristics of the motor system at different frequencies and accordingly proposes corresponding compensation strategies. For example, in the application scenario of a servo motor driving a robotic arm, when the robotic arm needs to move quickly and precisely to a specified position, the control system can adjust the operating parameters of the motor according to the real-time updated frequency response compensation sequence to ensure that each action can achieve the best effect. Then, the frequency response compensation sequence is optimized and reconstructed through adaptive filtering, and finally, a switching frequency adaptive mapping relationship is obtained. This adaptive filtering technology can dynamically adjust the filter parameters according to the actual operating conditions of the system, improving the stability and anti-interference ability of the system. Among them, the switching frequency adaptive mapping relationship includes key elements such as a frequency modulation matrix, a phase compensation vector, and a bandwidth optimization coefficient, which together constitute a complete control strategy framework to ensure that the motor system can operate efficiently and smoothly under various working conditions. To sum up, through the above series of complex but orderly data processing and technology applications, starting from the load characteristic parameter set, after a series of steps such as wavelet packet decomposition, Hilbert transform, and singular value decomposition, a switching frequency adaptive mapping relationship that accurately describes the dynamic behavior of the motor system is finally obtained. This achievement not only improves the performance of the motor control system but also provides solid technical support for realizing more intelligent and efficient industrial automation. Especially when dealing with complex and changeable working environments, such as various challenges encountered by a servo motor driving a robotic arm during the execution of diverse tasks, this method can help the control system adapt to changes more flexibly and ensure that each operation can be completed efficiently and precisely. Therefore, this series of technical means is not only an innovative breakthrough in theory but also an indispensable important link in practical engineering applications.

[0036] In a specific embodiment, the nonlinear optimization of a preset space vector division sector based on the switching frequency adaptive mapping relationship to obtain a corrected space vector positioning result includes: Conduct a topological structure analysis on the switching frequency adaptive mapping relationship to obtain a sector boundary feature sequence, and perform phase compensation on the sector boundary feature sequence through complex domain transformation to obtain a sector dynamic boundary matrix; Based on the sector dynamic boundary matrix, non-uniformly divide the preset space vector division sector to obtain a sector optimization distribution map, and perform boundary correction on the sector optimization distribution map through Lagrangian interpolation to obtain a sector correction parameter set; Perform vector space reconstruction on the sector correction parameter set to obtain a vector positioning feature group, and perform dynamic balance verification on the vector positioning feature group through Lyapunov stability analysis to obtain a vector positioning compensation sequence; Perform a non - linear mapping on the space vector based on the vector positioning compensation sequence to obtain a vector optimized positioning result, and perform phase synchronization on the vector optimized positioning result through hyperbolic function transformation to obtain a corrected space vector positioning result.

[0037] Specifically, in the high-frequency adaptive space vector pulse width modulation method, the process of nonlinearly optimizing the preset space vector division sectors based on the switching frequency adaptive mapping relationship to obtain the corrected space vector positioning result is a key step in achieving efficient motor control. First, this process begins with a topological structure analysis of the switching frequency adaptive mapping relationship to obtain the sector boundary feature sequence. At this stage, through a detailed analysis of the switching frequency adaptive mapping relationship, the optimal operating frequency and its corresponding sector boundary characteristics under different load conditions can be identified. Specifically, in the application scenario where a servo motor drives a robotic arm to perform a high-speed handling task, when the robotic arm grasps objects of different weights, the load on the motor will change accordingly, resulting in increased fluctuations in its internal electromagnetic torque. At this time, the sector boundary feature sequence extracted through topological structure analysis can help the control system more accurately understand the optimal sector division method under the current operating state. Next, phase compensation is performed on the sector boundary feature sequence through complex domain transformation to obtain the sector dynamic boundary matrix. Complex domain transformation is an effective mathematical tool that can adjust the phase information of a signal without changing its amplitude, which is crucial for improving the accuracy of space vector pulse width modulation. In the aforementioned application scenario, due to the change in load, a series of harmonic components will be generated by the fluctuations in the internal electromagnetic torque of the motor. These components not only affect the operating efficiency of the motor but may also cause distortion of the output waveform. Therefore, by performing phase compensation on the sector boundary feature sequence through complex domain transformation, a sector dynamic boundary matrix that more accurately describes the dynamic behavior of the motor system can be obtained, providing a basis for subsequent space vector division. Based on the sector dynamic boundary matrix, the preset space vector division sectors are non-uniformly divided to obtain a sector optimization distribution map, and Lagrange interpolation is used to correct the boundaries of the sector optimization distribution map to obtain a sector correction parameter set. This step aims to dynamically adjust the division method of space vectors according to the actual working conditions to make it more in line with the real-time requirements of the motor system. For example, in the application case of a servo motor driving a robotic arm, when the robotic arm needs to move quickly and accurately position, the traditional uniform sector division method may not meet the requirements. Non-uniform division can flexibly adjust the size and shape of each sector according to the actual load of the motor, ensuring that each sector can maximize its effectiveness. Subsequently, Lagrange interpolation technology is used to correct the boundaries of the sector optimization distribution map, further improving the accuracy and reliability of sector division. Then, vector space reconstruction is performed on the sector correction parameter set to obtain a vector positioning feature group, and Lyapunov stability analysis is used to perform dynamic balance verification on the vector positioning feature group to obtain a vector positioning compensation sequence. This step involves a comprehensive assessment of the stability of the motor system under different working conditions. Specifically, in the application scenario of a servo motor driving a robotic arm, as the load on the robotic arm changes, the stability of the motor system will also be affected.By dynamically balancing and verifying the vector positioning feature group, it is possible to ensure that the motor operates stably under various load conditions. The vector positioning compensation sequence provides the necessary adjustment strategies, enabling the motor system to maintain optimal performance in any situation. Finally, based on the vector positioning compensation sequence, a non-linear mapping of the space vector is performed to obtain the vector optimized positioning result, and the phase synchronization of the vector optimized positioning result is carried out through hyperbolic function transformation to obtain the corrected space vector positioning result. This step uses non-linear mapping technology to convert the vector positioning compensation sequence into specific control instructions to guide the precise operation of the power switching tubes in the inverter. For example, in the aforementioned application scenario, when the robotic arm performs high-precision assembly tasks, any slight position deviation may lead to the failure of the entire production process. Therefore, through the fine adjustment of the space vector by non-linear mapping and hyperbolic function transformation, it is ensured that each operation can achieve the expected effect. Among them, the corrected space vector positioning result includes key elements such as vector positioning coordinates, phase compensation angles, and positioning accuracy coefficients, which together constitute a complete control strategy framework to ensure that the motor system can operate efficiently and smoothly under various working conditions. In summary, through the above series of complex but orderly data processing and technology applications, starting from the switching frequency adaptive mapping relationship, through a series of steps such as topological structure analysis, complex domain transformation, and non-uniform partitioning, the corrected space vector positioning result that accurately describes the dynamic behavior of the motor system is finally obtained. This achievement not only improves the performance of the motor control system but also provides solid technical support for realizing more intelligent and efficient industrial automation. Especially when dealing with complex and changing working environments, such as the various challenges encountered by the servo motor-driven robotic arm during the execution of diverse tasks, this method can help the control system adapt to changes more flexibly and ensure that each operation can be completed efficiently and accurately. Therefore, this series of technical means is not only an innovative breakthrough in theory but also an indispensable important link in practical engineering applications.

[0038] In a specific embodiment, the obtaining of the high-frequency pulse width modulation waveform by performing dead-time adaptive compensation on the power switching tubes in the inverter according to the corrected space vector positioning result includes: Calculating the reference conduction time of the power switching tubes in the inverter based on the corrected space vector positioning result to obtain the reference switching timing, and performing dead-time pre-compensation on the reference switching timing to obtain the pre-compensated switching timing, where the pre-compensated switching timing includes the conduction and turn-off times of each power switching tube; Predicting the current ripple of the pre-compensated switching timing to obtain the current ripple prediction value, and performing non-linear adjustment on the pre-compensated switching timing based on the current ripple prediction value to obtain the adjusted switching timing; Perform voltage vector error compensation on the adjusted switching timing to obtain the compensated switching timing, and generate control signals for the inverter power switches based on the compensated switching timing, where the control signals include drive pulses for each power switch; Apply the control signals to the inverter power switches to obtain a high-frequency pulse width modulation waveform.

[0039] Specifically, in the high-frequency adaptive space vector pulse width modulation method, the process of performing dead-time adaptive compensation on the power switch tubes in the inverter according to the corrected space vector positioning result to obtain the high-frequency pulse width modulation waveform is one of the key steps to achieve efficient motor control. First, based on the corrected space vector positioning result, the reference conduction time of the power switch tubes in the inverter is calculated to obtain the reference switching sequence. This process involves accurately calculating the ideal conduction and turn-off moments of each power switch tube to ensure that the voltage vector output by the inverter can accurately follow the reference signal. For example, in the application scenario where a servo motor drives a robotic arm to perform a high-speed handling task, when the robotic arm grasps objects of different weights, the change in the motor load will cause an increase in current ripple, affecting the quality of the output waveform. By accurately calculating the reference switching sequence, a reliable starting point can be provided for subsequent optimization. Next, dead-time pre-compensation is performed on the reference switching sequence to obtain the pre-compensated switching sequence. The dead time is a time interval set to avoid short circuits caused by the simultaneous conduction of the upper and lower bridge arms. During this time interval, all power switch tubes are in the off state. However, this approach also introduces additional errors and affects the system performance. Therefore, by performing dead-time pre-compensation on the reference switching sequence, these problems can be alleviated to a certain extent. The pre-compensated switching sequence includes the specific conduction and turn-off times of each power switch tube. This step is crucial for reducing waveform distortion caused by the dead-time effect. Then, current ripple prediction is performed on the pre-compensated switching sequence to obtain the current ripple prediction value, and based on this prediction value, the pre-compensated switching sequence is non-linearly adjusted to obtain the adjusted switching sequence. Current ripple prediction is a complex but necessary step, which aims to identify and compensate for current fluctuations caused by the dead-time effect and other factors in advance. In the aforementioned application scenario, when the robotic arm needs to move quickly and accurately position, any small current fluctuation may affect the final position accuracy. By non-linearly adjusting the pre-compensated switching sequence, the switching moment can be dynamically optimized according to the actual operating conditions, further improving the system response speed and stability. Then, voltage vector error compensation is performed on the adjusted switching sequence to obtain the compensated switching sequence. Voltage vector error compensation is to correct the deviation between the output voltage and the ideal voltage caused by various reasons (such as dead time, switching delay, etc.). This step uses advanced algorithms and technologies to minimize the error and ensure that the voltage vector output by the inverter is as close as possible to the theoretical value. Based on the compensated switching sequence, control signals for the power switch tubes of the inverter are generated, where the control signals contain specific drive pulses for each power switch tube. These drive pulses directly determine the opening and closing timing of the power switch tubes and are crucial for achieving precise motor control. Finally, the control signals are applied to the power switch tubes of the inverter to obtain the high-frequency pulse width modulation waveform.Through the above series of complex processing steps, starting from the corrected space vector positioning result, after a series of operations such as reference conduction time calculation, dead time pre-compensation, current ripple prediction, and voltage vector error compensation, efficient inverter control is finally achieved. For example, in the application case of a servo motor driving a robotic arm, when the robotic arm is performing a high-precision assembly task, the high-frequency pulse width modulation waveform obtained by this method can not only ensure the efficient operation of the motor, but also significantly improve the stability and response speed of the system, enabling each operation to achieve the expected effect. Therefore, this series of technical means is not only an innovative breakthrough in theory, but also an indispensable important link in practical engineering applications, providing solid technical support for realizing more intelligent and efficient industrial automation.

[0040] In a specific embodiment, calculating a reference conduction time for the inverter power switch tubes based on the corrected space vector positioning result to obtain a reference switching timing includes: Determining a target voltage vector based on the corrected space vector positioning result, and performing a coordinate transformation on the target voltage vector to obtain a voltage vector in a two-phase stationary coordinate system; Judging the sector of the voltage vector in the two-phase stationary coordinate system to obtain the sector where the target voltage vector is located, and selecting a corresponding switching state vector based on the sector where the target voltage vector is located; Calculating the action time of the switching state vector to obtain the action time of each switching state vector, and calculating the conduction time of the power switch tubes based on the action time of each switching state vector and a preset switching frequency; Sorting the conduction time of the power switch tubes to obtain an initial reference switching timing, and calculating the duty cycle of each power switch tube by calculating the duty cycle of the initial reference switching timing; Generating a reference drive waveform for the power switch tubes based on the duty cycle, and modulating the reference drive waveform to obtain a reference switching timing.

[0041] Specifically, calculating the reference conduction time of the inverter power switch tubes based on the corrected space vector positioning result to obtain the reference switching sequence is one of the key steps to achieve efficient motor control. First, determine the target voltage vector based on the corrected space vector positioning result and perform coordinate transformation on it to obtain the voltage vector in the two-phase stationary coordinate system. This process involves converting the three-phase AC system into a two-phase rectangular coordinate system (αβ coordinate system) to more intuitively represent and process the voltage vector. For example, in the application scenario where a servo motor drives a robotic arm to perform a high-speed handling task, when the robotic arm grasps objects of different weights, the change in the motor load will cause the internal electromagnetic torque fluctuation to intensify. Coordinate transformation can more accurately describe these changes. In this process, the voltage vector in the two-phase stationary coordinate system includes the α-axis voltage component and the β-axis voltage component, which respectively represent the projections of the voltage vector in two orthogonal directions. Next, perform sector judgment on the voltage vector in the two-phase stationary coordinate system to determine the sector where the target voltage vector is located, and select the corresponding switching state vector based on this sector. Space vector pulse width modulation usually divides the plane into multiple sectors, and each sector corresponds to a specific combination of switching states. By judging the sector where the target voltage vector is located, the most appropriate switching state vector can be selected to approximate this voltage vector. This step is crucial for improving the quality of the inverter output waveform. For example, in the aforementioned application scenario, when the robotic arm needs to move quickly and accurately position, selecting the correct switching state vector can ensure that the torque output by the motor is sufficient and stable. Subsequently, calculate the action time of the switching state vector to obtain the action time of each switching state vector, and calculate the conduction time of the power switch tubes based on these action times and the preset switching frequency. This step involves complex mathematical operations and aims to determine the optimal conduction time of each power switch tube within a switching cycle to achieve optimal voltage vector synthesis. For example, adjusting the switching frequency according to the load demand can reduce harmonic distortion while ensuring efficiency. By accurately calculating the conduction time, the performance of the system can be significantly improved. Then, sort the conduction times of the power switch tubes to obtain the initial reference switching sequence, and calculate the duty cycle of this initial reference switching sequence to obtain the duty cycle of each power switch tube. The duty cycle refers to the ratio of the conduction time of the power switch tube within a switching cycle to the entire cycle, and it is an important parameter to measure the working state of the switch tube. For example, in the application case of a servo motor driving a robotic arm, when the robotic arm is performing a high-precision assembly task, accurately controlling the duty cycle of each power switch tube is crucial for ensuring the stable operation of the motor. Finally, generate the reference drive waveform of the power switch tubes based on the duty cycle and modulate this reference drive waveform to finally obtain the reference switching sequence. This step uses advanced modulation techniques to convert the theoretical calculation into an actual operable control signal to guide the accurate operation of the power switch tubes in the inverter.For example, through this modulation method, the operating state of the inverter can be dynamically adjusted according to the real-time load conditions to ensure that each operation can achieve the expected effect. The reference switching timing includes the specific conduction and turn-off moments of each power switch, which is of great significance for achieving efficient and stable motor control. In summary, through the above series of steps, starting from the corrected space vector positioning result, after a series of operations such as coordinate transformation, sector judgment, action time calculation, duty cycle calculation, etc., an efficient inverter control strategy is finally achieved, providing a solid technical support for realizing more intelligent and efficient industrial automation.

[0042] The high-frequency adaptive space vector pulse width modulation method in the embodiment of the present invention has been described above. Next, the high-frequency adaptive space vector pulse width modulation system in the embodiment of the present invention will be described. Please refer to Figure 2 , an embodiment of the high-frequency adaptive space vector pulse width modulation system in the embodiment of the present invention includes: An acquisition module 21, configured to acquire the real-time operation parameters of the motor system to obtain the motor phase current vector and the rotor position signal; An identification module 22, configured to identify the load characteristics of the motor system based on the motor phase current vector and the rotor position signal to obtain a set of load characteristic parameters; A deconstruction module 23, configured to perform frequency-domain deconstruction on the set of load characteristic parameters through multiple harmonic analyses to obtain a switching frequency adaptive mapping relationship; An optimization module 24, configured to perform non-linear optimization on the preset space vector division sectors based on the switching frequency adaptive mapping relationship to obtain a corrected space vector positioning result; A compensation module 25, configured to perform dead-time adaptive compensation on the switching sequence of the power switches in the inverter according to the corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform.

[0043] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the description in the above method embodiment, and details will not be repeated here.

[0044] Refer to Figure 3 , the embodiment of the present invention also provides a computer device, and its internal structure can be as Figure 3As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected via a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the above method is implemented.

[0045] Those skilled in the art can understand that Figure 3 the structure shown in is only a block diagram of some structures related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0046] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0047] Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0048] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article or method comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, apparatus, article or method. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, apparatus, article or method comprising such element.

[0049] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall equally be included in the patent protection scope of the present invention.

Claims

1. A high frequency adaptive space vector pulse width modulation method, characterized in that: The following steps are involved: Collect the real-time operating parameters of the motor system to obtain the motor phase current vector and rotor position signal; Performing load characteristic identification on the motor system based on the motor phase current vector and the rotor position signal to obtain a load characteristic parameter set; Deconstructing the load characteristic parameter set in the frequency domain through multiple harmonic analysis to obtain a switching frequency adaptive mapping relationship; Based on the switching frequency adaptive mapping relationship, nonlinear optimization is performed on the preset space vector division sectors to obtain a modified space vector positioning result; According to the modified space vector positioning result, the dead zone adaptive compensation is performed on the switching sequence of the power switch tubes in the inverter to obtain a high-frequency pulse width modulation waveform.

2. The high frequency adaptive space vector pulse width modulation method according to claim 1, characterized in that: The real-time operating parameters of the motor system are collected to obtain the motor phase current vector and the rotor position signal, including: Perform differential compensation sampling on the current sampling channels of the three-phase windings of the motor in the motor system to obtain an original three-phase current sequence, and perform two-dimensional coordinate mapping on the original three-phase current sequence through orthogonal decomposition technology to obtain a motor phase current vector; Performing complex vector decomposition on the motor phase current vector to obtain instantaneous vector components, and performing coordinate system transformation on the instantaneous vector components through synchronous rotation transformation to obtain a stator current space vector; The motor shaft position is sampled at high resolution by a preset rotor position encoder to obtain an initial mechanical angle sequence, and the initial mechanical angle sequence is subjected to magnetic pole logarithm analysis and phase calibration to obtain a rotor position signal.

3. The high frequency adaptive space vector pulse width modulation method according to claim 1, characterized in that: The step of identifying the load characteristics of the motor system based on the motor phase current vector and the rotor position signal to obtain a load characteristic parameter set includes: Decomposing the motor phase current vector in phase sequence to obtain an instantaneous current distribution sequence, and analyzing the power characteristics of the instantaneous current distribution sequence by cross-spectral analysis to obtain a load power spectrum density matrix; Calculating the motor load torque of the motor system based on the load power spectrum density matrix to obtain an instantaneous torque pulsation sequence, and reconstructing the spectrum of the instantaneous torque pulsation sequence through harmonic decomposition to obtain load dynamic characteristic parameters; Frequency domain feature extraction is performed on the load dynamic characteristic parameters to obtain a load frequency response curve, and parameter fitting is performed on the load frequency response curve through nonlinear mapping to obtain a load characteristic parameter set.

4. The high frequency adaptive space vector pulse width modulation method according to claim 1, characterized in that: The method of performing frequency domain deconstruction on the load characteristic parameter set by multiple harmonic analysis to obtain a switching frequency adaptive mapping relationship includes: Performing wavelet packet decomposition on the load characteristic parameter set to obtain a multi-scale frequency component sequence, and performing instantaneous phase extraction on the multi-scale frequency component sequence through Hilbert transform to obtain a harmonic feature vector group; Performing harmonic energy distribution analysis on the motor system based on the harmonic feature vector group to obtain a frequency response feature matrix, and performing feature space mapping on the frequency response feature matrix through a singular value decomposition technique to obtain a frequency domain feature distribution spectrum; Performing a hyperbolic slice analysis on the frequency domain characteristic distribution spectrum to obtain a frequency modulation curve family, and performing a nonlinear transformation on the frequency modulation curve family through a topological mapping technology to obtain a switching frequency mapping function; The dynamic frequency characteristics of the motor system are analyzed based on the switching frequency mapping function to obtain a frequency response compensation sequence, and the frequency response compensation sequence is optimized and reconstructed through adaptive filtering to obtain a switching frequency adaptive mapping relationship.

5. The high frequency adaptive space vector pulse width modulation method according to claim 1, characterized in that: The nonlinear optimization of the preset space vector division sectors based on the switching frequency adaptive mapping relationship to obtain a modified space vector positioning result includes: Performing a topological structure analysis on the switching frequency adaptive mapping relationship to obtain a sector boundary feature sequence, and performing phase compensation on the sector boundary feature sequence through complex domain transformation to obtain a sector dynamic boundary matrix; Based on the sector dynamic boundary matrix, the preset space vector division sector is non-uniformly divided to obtain a sector optimization distribution map, and the sector optimization distribution map is corrected by Lagrange interpolation to obtain a sector correction parameter set; Performing vector space reconstruction on the sector correction parameter set to obtain a vector positioning feature group, and performing dynamic balance verification on the vector positioning feature group through Lyapunov stability analysis to obtain a vector positioning compensation sequence; Based on the vector positioning compensation sequence, nonlinear mapping is performed on the space vector to obtain a vector optimization positioning result, and the vector optimization positioning result is phase synchronized through hyperbolic function transformation to obtain a modified space vector positioning result.

6. The high frequency adaptive space vector pulse width modulation method according to claim 1, characterized in that: The method of performing dead zone adaptive compensation on the power switch tube in the inverter according to the corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform includes: Based on the corrected space vector positioning result, a reference on-time calculation is performed on the power switch tube in the inverter to obtain a reference switch timing, and a dead time pre-compensation is performed on the reference switch timing to obtain a pre-compensated switch timing, wherein the pre-compensated switch timing includes the on-time and off-time of each power switch tube; Performing current ripple prediction on the pre-compensation switch timing to obtain a current ripple prediction value, and performing nonlinear adjustment on the pre-compensation switch timing based on the current ripple prediction value to obtain an adjusted switch timing; Performing voltage vector error compensation on the adjusted switch timing to obtain a compensated switch timing, and generating a control signal of an inverter power switch tube based on the compensated switch timing, wherein the control signal includes a drive pulse of each power switch tube; The control signal is applied to the inverter power switch tube to obtain a high-frequency pulse width modulation waveform.

7. The high frequency adaptive space vector pulse width modulation method according to claim 6, characterized in that: The step of calculating the reference on-time of the inverter power switch tube based on the corrected space vector positioning result to obtain the reference switch timing includes: Determining a target voltage vector based on the corrected space vector positioning result, and performing coordinate transformation on the target voltage vector to obtain a voltage vector in a two-phase stationary coordinate system; Performing sector judgment on the voltage vector in the two-phase stationary coordinate system to obtain the sector where the target voltage vector is located, and selecting a corresponding switch state vector based on the sector where the target voltage vector is located; Calculating the action time of the switch state vector to obtain the action time of each switch state vector, and calculating the conduction time of the power switch tube based on the action time of each switch state vector and a preset switching frequency; Sorting the on-time of the power switch tubes to obtain an initial reference switch timing sequence, and calculating the duty cycle of the initial reference switch timing sequence to obtain the duty cycle of each power switch tube; A reference driving waveform of the power switch tube is generated based on the duty cycle, and the reference driving waveform is modulated to obtain a reference switching timing.

8. A high frequency adaptive space vector pulse width modulation system, characterized in that: include: The acquisition module is used to collect the real-time operating parameters of the motor system to obtain the motor phase current vector and rotor position signal; an identification module, configured to identify the load characteristics of the motor system based on the motor phase current vector and the rotor position signal, and obtain a load characteristic parameter set; A deconstruction module, used for performing frequency domain deconstruction on the load characteristic parameter set through multiple harmonic analysis to obtain a switching frequency adaptive mapping relationship; An optimization module, configured to perform nonlinear optimization on the preset space vector division sectors based on the switching frequency adaptive mapping relationship to obtain a modified space vector positioning result; The compensation module is used to perform dead zone adaptive compensation on the switching sequence of the power switch tubes in the inverter according to the corrected space vector positioning result to obtain a high-frequency pulse width modulation waveform.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images