Intelligent control method and system of motor driver

Abstract

The invention discloses an intelligent control method and system for a motor driver, and the method comprises the steps: building an acoustoelectric incidence matrix through collecting harmonic current and sound pressure fluctuation data in the operation process of a motor, and forming a fault feature library; identifying an available fault mode to determine an energy collection point, and performing phase deviation processing on the collection point to generate an auxiliary driving signal; performing spectrum analysis on the sound pressure fluctuation data to obtain an acoustic characteristic spectrum, extracting an efficiency indication tone through characteristic decomposition, mapping the efficiency indication tone into a torque optimization parameter, and establishing a noise reduction control table; performing spectrum analysis on the enhanced control domain to identify a high-energy frequency band, executing a frequency shift operation to generate a safety frequency band signal, and forming a multi-frequency band management strategy in combination with a noise reduction control table; a management strategy is decomposed into parallel control flows for sequential staggered arrangement, and an energy-saving control matrix is generated; and generating a PWM control signal based on the matrix, carrying out edge reconstruction, and obtaining a softened edge waveform through an anti-resonance injection technology to complete intelligent control of the motor driver.

Description

Technical Field

[0001] The present invention relates to the technical field of motor drive control, and in particular to an intelligent control method and system for a motor driver. Background Art

[0002] Modern industrial motor systems commonly suffer from low efficiency and severe electromagnetic interference during operation. Traditional motor control technologies primarily rely on fixed-parameter control strategies, lacking real-time perception and adaptive adjustment capabilities for operating conditions. Existing fault detection methods often rely on analyzing single electrical signals, such as current and voltage. This fails to fully reflect the motor's true operating state and has limited ability to identify early-stage faults.

[0003] Furthermore, the steep switching signal edges generated by traditional PWM modulation techniques not only increase the power consumption of switching devices but also cause serious electromagnetic radiation issues. Existing energy-saving control strategies primarily rely on adjusting the voltage-frequency ratio, but lack in-depth analysis and optimal utilization of internal energy flows. Furthermore, traditional control methods ignore the impact of non-electrical information such as mechanical vibration and noise on system performance, and are unable to achieve the fusion and optimization of multi-physics information. Therefore, there is an urgent need to develop a motor drive technology that can comprehensively utilize multi-dimensional operational information, enabling intelligent perception and adaptive control. Summary of the Invention

[0004] The present invention provides an intelligent control method and system for a motor driver. The method aims to establish a comprehensive fault feature library by conducting in-depth feature mining and intelligent conversion of multi-source abnormal signals such as harmonic currents and sound pressure fluctuations during motor operation. This allows the identification of exploitable fault modes and energy recovery. Furthermore, the method extracts efficiency indicator sounds based on acoustic feature analysis, constructs a noise reduction control strategy through spectrum analysis and frequency shifting technology, and ultimately generates a PWM control signal with softened edge characteristics, thereby achieving efficient, low-noise, and intelligent control of the motor driver.

[0005] A first aspect of the present invention provides an intelligent control method for a motor driver, comprising the following steps: Collecting multi-source abnormal signals during motor operation, the multi-source abnormal signals including harmonic current and sound pressure fluctuation data, and performing feature conversion on the multi-source abnormal signals to form a fault feature library; Identifying an exploitable fault mode based on the fault feature library, performing energy recovery evaluation on the exploitable fault mode to determine an energy collection point, performing harmonic phase shift processing on the energy collection point to generate an auxiliary drive signal, and extracting amplitude-frequency characteristics of the auxiliary drive signal to determine an enhanced control domain; Performing spectral analysis on the sound pressure fluctuation data to generate an acoustic characteristic spectrum, performing feature decomposition on the acoustic characteristic spectrum to obtain an efficiency indicator sound, mapping the efficiency indicator sound into a torque optimization parameter, and adjusting the electromagnetic excitation frequency based on the torque optimization parameter to establish a noise reduction control table; Performing spectrum analysis on the enhanced control domain to identify a high-energy frequency band, performing a frequency shift operation on the high-energy frequency band to generate a safe frequency band signal, performing energy aggregation processing on the safe frequency band signal to form a frequency shift control sequence, and constructing a multi-band management strategy based on the fusion of the frequency shift control sequence and the noise reduction control table; Decomposing the multi-band management strategy into parallel control flows, staggering the parallel control flows in time sequence to generate a staggered control sequence, and generating an energy-saving control matrix based on the staggered control sequence; A PWM control signal is generated based on the energy-saving control matrix, edge reconstruction is performed on the PWM control signal to obtain a softened edge waveform, and the softened edge waveform is output to a power device to complete intelligent control of the motor driver.

[0006] A second aspect of the present invention provides an intelligent control system for a motor driver, comprising: A signal acquisition module is used to collect multi-source abnormal signals during the operation of the motor, the multi-source abnormal signals including harmonic current and sound pressure fluctuation data, and perform feature conversion on the multi-source abnormal signals to form a fault feature library; an energy recovery module, configured to identify exploitable fault modes based on the fault signature library, perform energy recovery evaluation on the exploitable fault modes to determine energy collection points, perform harmonic phase shift processing on the energy collection points to generate auxiliary drive signals, and extract amplitude-frequency characteristics of the auxiliary drive signals to determine an enhanced control domain; an acoustic optimization module, configured to perform spectral analysis on the sound pressure fluctuation data to generate an acoustic characteristic spectrum, perform feature decomposition on the acoustic characteristic spectrum to obtain an efficiency indicator sound, map the efficiency indicator sound into a torque optimization parameter, and adjust the electromagnetic excitation frequency based on the torque optimization parameter to establish a noise reduction control table; a frequency management module configured to perform spectrum analysis on the enhanced control domain to identify high-energy frequency bands, perform frequency shifting on the high-energy frequency bands to generate safe frequency band signals, perform energy aggregation processing on the safe frequency band signals to form a frequency shift control sequence, and construct a multi-band management strategy based on the fusion of the frequency shift control sequence and the noise reduction control table; a timing control module, configured to decompose the multi-band management strategy into parallel control flows, perform time-staggered arrangement on the parallel control flows to generate an interleaved control sequence, and generate an energy-saving control matrix based on the interleaved control sequence; An output shaping module is used to generate a PWM control signal based on the energy-saving control matrix, perform edge reconstruction on the PWM control signal to obtain a softened edge waveform, and output the softened edge waveform to the power device to complete the intelligent control of the motor driver.

[0007] The beneficial effects of the present invention are reflected in the following points: First, through the multi-source abnormal signal fusion analysis technology, the harmonic current and sound pressure fluctuation signals are constructed into an acoustic-electric correlation matrix, which realizes the accurate identification of the fault mode, and converts the identified fault energy into an auxiliary drive signal, thereby enhancing the driving ability of the system and improving the accuracy of fault diagnosis. Secondly, by extracting the efficiency indicator sound from the acoustic characteristic spectrum, a direct correlation between the acoustic signal and the motor efficiency state is established. The efficiency indicator sound can reflect the energy conversion efficiency of the motor in real time. By mapping it into a torque optimization parameter and adjusting the electromagnetic excitation frequency, the motor efficiency optimization control based on acoustic feedback is realized, thereby improving the motor operation efficiency. Finally, the frequency shifting technology is used to transfer the high-energy interference frequency band to a safe area, and the parallel control flow with time-staggered timing is combined to reduce the concentration of switching losses, and the anti-resonance injection technology is used to soften the edge of the PWM signal, which effectively reduces the electromagnetic interference level and switching power consumption, and improves the electromagnetic compatibility and energy consumption control effect of the system.

[0008] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The accompanying drawings herein illustrate specific examples of the technical solutions described in the present invention, and together with the specific implementation methods constitute a part of the specification, and are used to explain the technical solutions, principles and effects of the present invention.

[0010] Unless otherwise specified, the same reference numerals in different drawings represent the same or similar technical features, and the same or similar technical features may also be represented by different reference numerals.

[0011] Figure 1 It is a flow chart of an intelligent control method for a motor driver of the present invention.

[0012] Figure 2 It is a structural block diagram of an intelligent control system of a motor driver of the present invention. DETAILED DESCRIPTION

[0013] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, circuits, and

[0014] It is to be understood that the terminology "including", "comprising", "consisting" and "consisting essentially of" used in the specification and the appended claims, are used in the sense of making an inclusion of the features, integers, steps, operations, elements, and / or components listed thereafter, but does not preclude the inclusion of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0015] Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" or "in other embodiments" or "in still other embodiments" in various places throughout this specification are not necessarily all referring to the same embodiment, unless otherwise specified. The terms "including", "containing", "having" and variations thereof mean "including but not limited to", unless expressly specified otherwise.

[0016] The technical solutions of the embodiments of the present application are introduced as follows.

[0017] As shown in Figure 1 The intelligent control method of the motor driver provided by the embodiments of the present application includes the following steps S110-S160: In step S110, multi-source abnormal signals in the motor running process are collected, the multi-source abnormal signals include harmonic current and sound pressure fluctuation data, and the multi-source abnormal signals are subjected to feature conversion to form a fault feature library.

[0018] Specifically, the system collects multi-source abnormal signals during motor operation. High-precision current sensors are installed on each phase of the motor's stator winding to monitor current waveform changes in real time. Harmonic currents are collected using Rogowski coil current sensors with a frequency response range from DC to 20 kHz, accurately capturing fundamental and higher harmonic components. The sensors continuously record three-phase current signals at a sampling frequency of 50 kHz, achieving 16-bit data accuracy. Simultaneously, an acoustic sensor array is placed on the motor housing to collect acoustic pressure fluctuations generated by the motor's operation. These acoustic sensors use MEMS microphones with a frequency response range of 20 Hz to 20 kHz and a dynamic range greater than 100 dB. The sensor array consists of eight microphones arranged in a regular octagon around the motor, maintaining a fixed distance from the motor surface. Acoustic pressure fluctuation data is collected synchronously at 48 kHz and time-stamped with the current signals. The acquisition system is equipped with a high-speed data buffer, enabling 48 hours of uninterrupted recording. Multi-source abnormal signal data includes four dimensions: time stamp, sensor location, signal amplitude, and frequency characteristics. Anti-aliasing filtering and noise suppression are implemented during the acquisition process to ensure signal quality. The raw data is pre-processed and stored in a unified format.

[0019] In some embodiments, the feature conversion of the multi-source abnormal signals to form a fault feature library includes: constructing a current distortion spectrum based on the harmonic current; performing inter-cycle difference calculation on the current distortion spectrum to identify the distortion growth rate; constructing an acoustic-electric correlation matrix based on the distortion growth rate and the sound pressure fluctuation data; and establishing a fault feature library based on the acoustic-electric correlation matrix.

[0020] A current distortion spectrum is constructed based on harmonic currents. The collected three-phase current signals are subjected to a fast Fourier transform to extract the amplitude and phase information of each harmonic component. The harmonic analysis window length is set to an integer multiple of the fundamental period to ensure accurate spectral resolution. The total harmonic distortion (THD) is calculated as √(∑I_h²) / I_1, where I_h is the effective value of the hth harmonic and I_1 is the effective value of the fundamental. The THD values ​​of each phase current are arranged in chronological order to form a distortion time series curve. The time series curve is mapped two-dimensionally, with time on the horizontal axis and frequency on the vertical axis, and color depth indicating distortion intensity, to construct a current distortion spectrum. The spectrum uses wavelet transform for time-frequency analysis, with the Morlet wavelet selected as the window function and a scale parameter covering the frequency range of 1-100 Hz. A continuous wavelet transform is used to obtain a high-resolution time-frequency distribution, highlighting transient distortion characteristics. Different colored areas in the distortion map correspond to different motor failure modes. For example, a winding short circuit manifests as increased high-frequency distortion, while bearing wear manifests as periodic distortion at specific frequencies. The map data is organized in a matrix format with dimensions equal to the number of time points x the number of frequency points. Each element records the distortion intensity at the corresponding time-frequency point.

[0021] The current distortion spectrum is segmented by cycle-to-cycle differences to identify the distortion growth rate. The current distortion spectrum is segmented according to the motor fundamental wave period, with each cycle corresponding to a time segment in the spectrum. The distortion difference between corresponding frequency points between adjacent cycles is calculated: ΔD(f,t)=D(f,t+T)-D(f,t), where D(f,t) is the distortion intensity at frequency f at time t, and T is the fundamental wave period. Statistical analysis is performed on the difference matrix to calculate the distortion growth rate for each frequency component: G(f)=mean(ΔD(f,t)) / std(ΔD(f,t)), where mean represents the mean and std represents the standard deviation. A growth rate greater than zero indicates an increasing distortion trend at that frequency, while a growth rate less than zero indicates a decreasing distortion trend. A sliding average filter is used to eliminate the influence of random fluctuations and extract a stable growth trend. The growth rates for each frequency are organized into a growth rate vector, with a length equal to the number of frequency resolution points. The growth rate vector is normalized to eliminate the influence of magnitude differences between frequency components. Identify frequency intervals with abnormal growth rates, which typically correspond to the development stages of internal motor faults. The distortion growth rate not only reflects the deterioration trend of the motor's electrical performance but also provides important information for predicting the timeline of fault evolution.

[0022] An acoustic-electric correlation matrix is ​​constructed based on the distortion growth rate and sound pressure fluctuation data. The sound pressure fluctuation data is decomposed in the frequency domain to extract the acoustic characteristic frequencies corresponding to the current harmonics. The sound pressure signal is processed using a short-time Fourier transform, with the window length consistent with the current analysis to ensure the accuracy of the time-frequency correspondence. The gradient of the sound pressure amplitude over time is calculated to identify the abrupt changes and development trends of the acoustic signal. Correlation analysis is performed between the distortion growth rate vector and the sound pressure gradient vector, and the Pearson correlation coefficient matrix R is calculated, where R(i,j) represents the correlation between the i-th current frequency and the j-th acoustic frequency. The correlation coefficient is calculated using a sliding window method with a window size of 10 fundamental wave periods and a 50% overlap between windows. The acoustic-electric correlation matrix C is constructed, with the matrix element C(i,j) = R(i,j) × G(i) × A(j), where G(i) is the distortion growth rate of current frequency i and A(j) is the amplitude weight of acoustic frequency j. The correlation matrix reveals the coupling relationship between electromagnetic and mechanical faults in the motor, with high-correlation regions indicating the co-evolution of multi-source faults. Principal component analysis was used to reduce the dimensionality of the correlation matrix and extract the primary acoustic-electrical correlation patterns, retaining over 95% of the variance contribution.

[0023] A fault feature library is established based on the acoustic-electrical correlation matrix. Fault types include five major categories: stator winding fault, rotor bar fracture, bearing defect, air gap eccentricity, and load imbalance. For each fault type, the characteristic correlation matrix pattern is extracted, including statistical features such as the matrix's main diagonal distribution, peak location, and energy concentration area. A clustering algorithm is used to perform unsupervised grouping of the correlation matrices of historical fault samples to identify typical fault feature patterns. A feature similarity metric function is established, using the matrix binorm to calculate the distance between different correlation matrices. A hierarchical feature index structure is constructed to support fast fault pattern matching and retrieval. The fault feature library adopts a relational database architecture, with each record containing fields such as the fault type label, correlation matrix data, fault severity, and occurrence time. The feature library supports incremental collection: new fault samples are automatically added to the library after feature extraction, continuously enriching the coverage of fault modes. Regular feature library maintenance removes redundant features with high repetition rates, keeping the feature library streamlined and efficient.

[0024] Step S120 , identifying available fault modes based on the fault feature library, performing energy recovery evaluation on the available fault modes to determine energy collection points, performing harmonic phase shift processing on the energy collection points to generate auxiliary drive signals, and extracting amplitude-frequency characteristics of the auxiliary drive signals to determine enhanced control domains.

[0025] Specifically, exploitable fault modes are identified based on a fault signature database. A multi-metric comprehensive evaluation approach is employed, with key evaluation indicators including fault energy density, frequency stability, spatial concentration, and temporal persistence. The energy density metric E_density = ∑P_harm / A_region is calculated for each fault mode, where E_density represents energy density, P_harm represents the sum of all harmonic powers, and A_region represents the area affected by the fault. Frequency stability is measured by the standard deviation of the harmonic frequencies; fault modes with high stability are more suitable for energy harvesting. Spatial concentration assesses the concentration of fault energy within the motor's spatial distribution; high concentration indicates higher energy harvesting efficiency. Temporal persistence analyzes the development cycle of the fault mode; long-lasting faults offer more energy recovery opportunities. Suitable fault modes are screened by setting appropriate thresholds, which are dynamically adjusted based on motor type and operating conditions. Identified exploitable fault modes primarily include periodic impact energy caused by bearing rolling element defects, magnetic field harmonics generated by stator slot harmonics, air gap magnetic field nonuniformity caused by rotor eccentricity, and torque ripple energy caused by load fluctuations. A detailed energy characteristic file is established for each exploitable fault mode, recording key information such as its frequency range, power level, occurrence location, and evolution trend.

[0026] In some embodiments, the energy recovery evaluation of the exploitable fault mode to determine the energy collection point includes: extracting the energy distribution of the harmonic current from the exploitable fault mode; determining a resonance area based on the energy distribution; performing energy density evaluation on the resonance area to generate a high-density collection area; and determining an energy collection point based on the high-density collection area.

[0027] The energy distribution of harmonic currents is extracted from exploitable fault modes. A windowed Fourier transform is used to perform time-frequency analysis on the current signal under exploitable fault modes. Integer multiples of the fundamental period are used as the window length, and an appropriate overlap ratio is set to obtain an energy distribution map with high time-frequency resolution. The instantaneous power of each harmonic is calculated as P_h(t) = I_h(t) × V_h(t) × cos(φ_h), where P_h(t) is the instantaneous power of the hth harmonic, I_h(t) and V_h(t) are the effective current and voltage values ​​of the hth harmonic, respectively, and φ_h is the corresponding phase angle. The instantaneous power is mapped three-dimensionally according to spatial position and frequency dimensions to construct a stereoscopic energy distribution map. The spatial dimension uses a cylindrical coordinate system to describe the internal position information of the motor, while the frequency dimension covers the range from the fundamental to higher harmonics. An interpolation algorithm is used to fill in the energy values ​​between discrete measurement points to generate a continuous energy distribution function. The energy distribution map is represented by isoenergy lines, whose density reflects the steepness of the energy gradient. Identify the main characteristics of energy distribution, including key information such as energy peak location, energy propagation path, and energy attenuation pattern. Build a statistical model of energy distribution and calculate statistical parameters such as energy mean, variance, skewness, and kurtosis.

[0028] Exemplarily, determining the resonance area based on the energy distribution includes: evaluating the positioning complexity based on the energy distribution to determine the search accuracy, the positioning complexity including the number of energy peaks, distribution discreteness and harmonic order; setting area division parameters according to the search accuracy; and using the area division parameters to perform spatial mapping processing on the energy distribution to generate a resonance area.

[0029] Search accuracy is determined by evaluating positioning complexity based on energy distribution. Complexity analysis is performed on the obtained energy distribution function to quantify the difficulty of locating the resonance region. The number of energy peaks is counted using a local maximum detection algorithm. After removing noise using a morphological opening operation, the number of points that meet the local maximum condition is calculated. Distribution dispersion is measured using the coefficient of variation: CV = σ_E / μ_E, where CV is the coefficient of variation, σ_E is the standard deviation of the energy distribution, and μ_E is the mean of the energy distribution. A larger CV value indicates a more uneven energy distribution. Harmonic order complexity is characterized by the number of effective harmonics, counting the number of harmonic components whose amplitude exceeds a certain proportion of the maximum harmonic. A comprehensive positioning complexity index is established, combining the number of energy peaks, distribution dispersion, and harmonic order in a weighted combination. The weight coefficient is determined through empirical statistical data. Search accuracy is determined based on the complexity level. A coarse search strategy is used for low complexity and a fine search strategy for high complexity. Search accuracy parameters include spatial resolution, angular resolution, axial resolution, and frequency resolution. A mapping relationship table between precision and complexity is established to achieve adaptive adjustment of search precision and improve the efficiency and accuracy of resonance region identification.

[0030] Set the region partitioning parameters based on the search accuracy. Based on the determined search accuracy, design a spatial region partitioning strategy. Radial partitioning uses a constant-step method, determining the appropriate partitioning interval and number of layers based on the search accuracy requirements. Angular partitioning takes into account the symmetry of the motor, and the partitioning interval is typically an integer multiple of the number of motor pole pairs to ensure reasonable partitioning. Axial partitioning is determined based on the motor length to ensure coverage of the entire effective length range of the motor. Frequency partitioning uses a logarithmic spacing strategy, with dense partitioning in the low-frequency band and sparse partitioning in the high-frequency band, balancing computational accuracy and efficiency. Establish a four-dimensional grid coordinate system, with each grid node corresponding to an energy sampling point. Design a grid optimization algorithm to increase computational efficiency by increasing the density of grid points in regions with large energy gradients and reducing the density of grid points in regions with flat energy. Determine the threshold parameter for region merging, automatically merging adjacent regions when the energy density difference is less than the set threshold. Establish a smoothing method for region boundaries, using an interpolation algorithm to eliminate jagged boundaries and obtain smooth region contours.

[0031] Resonance regions are generated by spatially mapping the energy distribution using the region partitioning parameters. A four-dimensional grid coordinate system is established based on the specified region partitioning parameters. The energy distribution space is gridded according to the radial partitioning interval Δr, the angular partitioning interval Δθ, the axial partitioning interval Δz, and the frequency partitioning interval Δf. The continuous energy distribution function is discretized into a grid data structure, with each grid node corresponding to an energy sampling point. The energy integral is calculated for each grid cell according to the accuracy requirements of the region partitioning parameters, using a high-precision numerical integration method to improve accuracy. A region growing algorithm is used, starting from a seed point with high energy density, to expand the growing boundary toward the surrounding region based on a merging threshold specified by the region partitioning parameters. The growing criterion is based on the energy density difference and gradient direction consistency of adjacent cells. Regions are merged when the difference is less than the threshold specified in the region partitioning parameters. Morphological post-processing is performed on the grown regions, including hole filling, boundary smoothing, and noise removal. Shape descriptors are calculated for each region, including geometric features such as compactness, slenderness ratio, and convex hull ratio. Anomalous regions are filtered out based on the shape constraints of the region partitioning parameters. The resulting resonance region constitutes the spatial basis for energy collection and provides an accurate candidate range for further screening of high-density collection areas.

[0032] Energy density is assessed in the resonant region to generate high-density collection zones. The resonant region is divided into several subregions, each with its own size determined based on the expected size of the energy harvesting device. The average energy density of each subregion is calculated as ρ_E = E_total / V_region, where ρ_E is the energy density, E_total is the total energy within the subregion, and V_region is the subregion volume. Kernel density estimation is used to smooth the energy density distribution, employing a Gaussian kernel function and optimizing the bandwidth parameters through cross-validation. An energy density threshold is established, and subregions exceeding the threshold are marked as high-density regions. The threshold is determined by considering the minimum operating power and economic requirements of the energy harvesting device. High-density regions are merged, with adjacent regions with similar energy densities being combined into a large collection zone. The energy quality factor of each high-density collection zone is calculated, taking into account factors such as energy density, collection area, and energy duration. Collection zones are prioritized, with high-priority areas selected as preferred energy harvesting locations. Practical considerations for energy harvesting are considered, including device installation ease, maintenance accessibility, and impact on motor operation. Establish a database of high-density collection areas and record detailed information such as the location coordinates, energy parameters, geometric dimensions, and access methods of each collection area.

[0033] Energy collection points are identified based on high-density collection areas. A multi-objective optimization approach is chosen, with optimization objectives including maximizing energy density, optimizing device matching, and minimizing installation complexity. A multi-objective optimization function is established, comprehensively considering various performance indicators and constraints. An intelligent optimization algorithm is used to solve the multi-objective optimization problem, with appropriate population size and iteration parameter settings. Pareto optimal solutions are clustered and cluster centers are selected as candidate energy collection points. Considering the mutual influence between multiple collection points, graph theory methods are used to analyze the topology of the collection point network to ensure network connectivity and robustness. The final number of energy collection points is determined based on system requirements and cost constraints, forming a reasonable distributed network. A dedicated interface circuit is designed for each energy collection point, including an impedance matching network, power conditioning circuit, and signal isolation module. A monitoring system for energy collection points is established to monitor the energy output and operating status of each point in real time to ensure stable system operation.

[0034] Harmonic phase shifting is performed on energy collection points to generate auxiliary drive signals. Harmonic current phase information is extracted from each energy collection point, and the phase difference between them (Δφ_ij = φ_i - φ_j) is calculated, where φ_i and φ_j represent the phases of the i-th and j-th collection points, respectively. A phase difference matrix is ​​then constructed. Phase shifting is used to adjust the signal phases at each collection point, allowing multiple signals to superimpose in phase and increasing the amplitude of the composite signal. Phase shifting utilizes digital signal processing methods, using the Hilbert transform to extract the instantaneous phase of the signal. A phase compensation filter is then used to achieve precise phase adjustment. A phase compensation function is designed to correct the phases of different frequency components. An optimization algorithm is used to determine the optimal phase shift parameters to maximize the power density of the processed signal. The phase-shifted signal passes through a power synthesis circuit to generate a unified auxiliary drive signal. This auxiliary drive signal has enhanced amplitude and improved waveform quality, providing additional drive capability for the motor control system. An adaptive weighting strategy is employed in the signal synthesis process, dynamically adjusting the synthesis weights based on the signal quality of each collection point. The generated auxiliary drive signal is isolated from the main drive circuit through an isolation amplifier to ensure the safety and stability of the system.

[0035] The auxiliary drive signal's amplitude-frequency characteristics are extracted to determine the enhanced control domain. A high-resolution fast Fourier transform is used to extract the signal's frequency domain characteristics, and a Hanning window function is used as the analysis window to reduce spectral leakage. The signal's amplitude and phase responses at each frequency point are calculated, and a complete Bode plot is plotted. The auxiliary drive signal's primary frequency components and harmonic distribution are identified, and the signal's effective bandwidth and power spectral density are determined. Bandpass filtering is used to extract frequency components beneficial to motor control and filter out unwanted noise and interference. The auxiliary drive signal's dynamic characteristics, including response speed, steady-state accuracy, and interference rejection, are analyzed. A transfer function model of the auxiliary drive signal is established, and the model parameters are obtained through system identification methods. Controller parameters are designed based on the transfer function model to achieve coordinated coordination between the auxiliary drive signal and the main control system. The boundary conditions of the enhanced control domain, including the frequency range, amplitude range, and phase range, are determined. The enhanced control domain defines the parameter space within which the auxiliary drive signal can effectively operate, within which the auxiliary drive signal can significantly improve the motor's control performance.

[0036] Step S130 , performing spectral analysis on the sound pressure fluctuation data to generate an acoustic characteristic spectrum, performing feature decomposition on the acoustic characteristic spectrum to obtain an efficiency indicator sound, mapping the efficiency indicator sound to a torque optimization parameter, and adjusting the electromagnetic excitation frequency based on the torque optimization parameter to establish a noise reduction control table.

[0037] Specifically, the sound pressure fluctuation data is spectrally analyzed to generate an acoustic signature spectrum. The collected sound pressure fluctuation data is input into a high-resolution spectrum analysis system for time-domain preprocessing, including DC component removal and preliminary filtering. The sound pressure fluctuation data is segmented using a longer window length, a Hanning window function is used to reduce spectral leakage, and a high window overlap ratio is set to ensure spectral continuity. The power spectral density (PSD(f)) of the sound pressure fluctuation data is calculated as |X(f)|² / fs, where PSD(f) is the power spectral density at frequency f, X(f) is the Fourier transform of the sound pressure fluctuation data, fs is the sampling frequency, and |X(f)|² represents the square of the frequency domain signal amplitude. The power spectral density of the sound pressure fluctuation data is logarithmically transformed to enhance the discernibility of weak frequency components. Multi-resolution analysis technology is employed to analyze the sound pressure fluctuation data at different resolutions in different frequency bands: high resolution is used for low-frequency bands and moderate resolution is used for high-frequency bands, balancing analysis accuracy and computational efficiency. Spectral subtraction is used to remove the influence of background noise from the sound pressure fluctuation data and extract pure acoustic signal components. The main frequency components in the sound pressure fluctuation data spectrum are identified, including the fundamental frequency, harmonic frequencies, modulation frequency, and resonant frequency. A two-dimensional frequency-amplitude feature map based on the sound pressure fluctuation data is constructed, with the horizontal axis representing frequency, the vertical axis representing sound pressure amplitude, and the color depth representing energy density. A joint time-frequency analysis is performed on the characteristic spectrum of the sound pressure fluctuation data, using the short-time Fourier transform to capture the time-varying characteristics of the acoustic signal.

[0038] In some embodiments, the feature decomposition of the acoustic characteristic spectrum to obtain the efficiency indicator sound includes: performing frequency component separation on the acoustic characteristic spectrum to extract efficiency-related components; performing feature screening processing on the efficiency-related components to generate a dominant frequency set; performing correlation analysis on the dominant frequency set and the harmonic current to obtain efficiency characteristic values; and extracting the efficiency indicator sound based on the efficiency characteristic values.

[0039] The acoustic characteristic spectrum is separated by frequency components to extract efficiency-related components. The acoustic characteristic spectrum is input into the frequency component separation system, and wavelet decomposition technology is used to decompose the complex spectral signal into components of different frequency scales. Appropriate wavelet basis functions are selected and the appropriate number of decomposition layers is set to cover the entire frequency band from low to high frequencies. Each decomposition layer corresponds to a specific frequency range, with lower layers corresponding to high-frequency components and higher layers corresponding to low-frequency components. The energy distribution of each frequency component is calculated as E_i=∫|W_i(f)|²df, where E_i is the energy value of the i-th layer, W_i(f) is the wavelet decomposition coefficient of the i-th layer, and the integral symbol represents the integral operation over the entire frequency range. Energy threshold screening is used to retain frequency components with an energy proportion exceeding a certain proportion of the total energy. Empirical mode decomposition is used to further refine the frequency components, decomposing each major frequency band into several intrinsic mode functions. The instantaneous frequency and instantaneous amplitude of each intrinsic mode function are analyzed to identify frequency components that are synchronized with changes in motor efficiency. Establish efficiency correlation criteria and identify efficiency-related components by analyzing the correlation between frequency components and efficiency-influencing factors such as motor load, speed, and temperature. Use correlation analysis to calculate the Pearson correlation coefficient between each frequency component and the efficiency parameter. Components with an absolute value greater than a set threshold are selected as efficiency-related components. Reconstruct these efficiency-related components to form a pure efficiency-related acoustic signal. Build a feature library for efficiency-related components, recording each component's attributes such as frequency range, energy characteristics, and time-varying properties.

[0040] Feature screening was performed on efficiency-related components to generate a dominant frequency set. A frequency importance evaluation index was established, comprehensively considering the energy intensity, stability, and discrimination of frequency components. The signal-to-noise ratio (SNR) of each frequency component was calculated, prioritizing frequency components with high SNRs. The temporal stability of the frequency components was analyzed, the standard deviation of frequency drift was calculated, and components with stable frequencies were selected. The ability of frequency components to discriminate between different efficiency levels was assessed, and the inter-class separation of each component was calculated using Fisher discriminant analysis. A multi-criteria decision model was established to comprehensively score the frequency components. The scoring function was S_i = w1 × SNR_i + w2 × Stability_i + w3 × Discrimination_i, where S_i is the comprehensive score of the i-th frequency component, SNR_i is the signal-to-noise ratio, Stability_i is the stability index, Discrimination_i is the discrimination index, and w1, w2, and w3 are the corresponding weight coefficients. Frequency components were ranked according to their scores, and the top-ranked components were selected to form the dominant frequency set. Cluster analysis was performed on the dominant frequency set to identify and merge similar frequency components.

[0041] Correlation analysis is performed between the dominant frequency set and the harmonic current to obtain efficiency eigenvalues. Each harmonic component of the harmonic current is extracted, and the effective value and phase angle of each harmonic are calculated. An acoustic-electrical feature matching matrix is ​​established, with the matrix elements representing the correlation coefficients between the dominant frequency and the harmonic current components. A sliding window correlation analysis method is used to calculate the correlation variation within different time windows, with the window length set to an integer multiple of the fundamental period. The correlation coefficient matrix R_ij = corr(A_i, I_j) is calculated, where R_ij represents the correlation coefficient between the i-th dominant frequency and the j-th harmonic current, A_i represents the amplitude sequence of the i-th dominant frequency, I_j represents the amplitude sequence of the j-th harmonic current, and corr represents the correlation coefficient calculation function. Highly correlated acoustic-electrical feature pairs are identified, and feature combinations with absolute correlation coefficients greater than a set threshold are selected. The time-varying nature of the correlation is analyzed, and a dynamic time warping algorithm is used to address the temporal alignment of different features. A multivariate regression model is established, with the dominant frequency set as the independent variable and motor efficiency as the dependent variable. The regression coefficients represent the efficiency eigenvalues. A stepwise regression method was used to screen for acoustic features that significantly impacted efficiency, eliminating frequency components with low contributions. The partial correlation coefficient for each feature was calculated to assess its independent impact on efficiency while controlling for other variables. An efficiency eigenvalue vector was constructed, containing the efficiency impact weights corresponding to each dominant frequency. The efficiency eigenvalues ​​were normalized to eliminate the influence of different feature dimensions.

[0042] The efficiency indicator sound is extracted based on the efficiency eigenvalue. Using the efficiency eigenvalue as a weight coefficient, a weighted synthesis is performed on the dominant frequency set: EIS(t) = ∑λi × A_i(t) × cos(2πf_i × t + φ_i), where EIS(t) is the efficiency indicator sound signal at time t, λi is the efficiency eigenvalue (weight) corresponding to the i-th dominant frequency, A_i(t) is the time-varying amplitude of the i-th dominant frequency at time t, f_i is the i-th dominant frequency value, φ_i is the phase angle of the i-th dominant frequency, and ∑ represents the sum of all dominant frequency components. Noise components are removed from the efficiency indicator sound through filtering, and a bandpass filter is used to retain the signal within the effective frequency range. Envelope detection is performed on the efficiency indicator sound to extract the signal's envelope characteristics. Envelope variations reflect fluctuations in motor efficiency. Statistical characteristics of the efficiency indicator sound, including mean, standard deviation, crest factor, and skewness, are calculated to establish a quantitative indicator of efficiency status. A quality assessment method for the efficiency indicator sound is designed, and the accuracy of the sound is verified by comparing it with actual efficiency measurements. Establish a threshold system for efficiency indicator sounds and define the characteristic ranges of indicator sounds corresponding to different efficiency levels. Conduct time-frequency analysis of the efficiency indicator sounds to observe their variations under different operating conditions. Establish a database of efficiency indicator sounds to record the characteristics of the indicator sounds under different motor types and operating conditions. As an acoustic indicator of the motor's efficiency status, the efficiency indicator sound can monitor changes in the motor's energy conversion efficiency in real time. When factors such as magnetic circuit saturation, increased mechanical friction, and winding heating occur within the motor, the frequency components and amplitude characteristics of the efficiency indicator sound will change accordingly. By monitoring the changing trends of the efficiency indicator sound, early signs of motor efficiency degradation can be identified in a timely manner.

[0043] The efficiency indicator sound is mapped into torque optimization parameters. The frequency components of the efficiency indicator sound are directly correlated with the electromagnetic torque ripple of the motor. By analyzing this correlation, the torque optimization adjustment parameters can be derived. A torque optimization function is established, taking the amplitude, dominant frequency, and phase characteristics of the efficiency indicator sound as input variables. The least squares method is used to fit the parameters of the mapping function to ensure the accuracy of the mapping relationship. The correlation between the amplitude variation of the efficiency indicator sound and the torque ripple coefficient is analyzed, and an amplitude-ripple mapping model is established. The relationship between the frequency drift of the efficiency indicator sound and the torque output stability is studied to determine the sensitive range for frequency adjustment. The phase information of the efficiency indicator sound is calculated, and the phase characteristics are converted into torque phase compensation parameters. A multidimensional torque optimization parameter vector is established, including key parameters such as the amplitude adjustment coefficient, frequency compensation amount, and phase correction angle. The effectiveness of the mapping relationship is verified through numerical simulation to ensure that the torque optimization parameters accurately reflect the operating characteristics of the motor. An adaptive mapping mechanism is established to consider the changes in the mapping relationship under different operating conditions.

[0044] A noise reduction control table is established by adjusting the electromagnetic excitation frequency based on torque optimization parameters. The obtained torque optimization parameters are used as input to design an adjustment strategy for the electromagnetic excitation frequency. The goal of adjusting the electromagnetic excitation frequency is to minimize acoustic noise while ensuring motor performance. A frequency adjustment algorithm is developed to calculate the excitation frequency offset for each phase winding based on the torque optimization parameters. A frequency adjustment function is designed to establish a mathematical relationship between the torque optimization parameters and the frequency offset. Frequency adjustment utilizes a multi-frequency coordinated control strategy, simultaneously optimizing the fundamental frequency and harmonic frequencies. The impact of frequency adjustment on the motor torque output is calculated to ensure that the adjusted frequency does not significantly reduce motor efficiency. Constraints are established, including the frequency adjustment range, torque ripple limit, and efficiency loss threshold. An optimization algorithm is used to determine the optimal frequency adjustment solution under these multiple constraints. The optimization results are organized into a noise reduction control table containing fields such as operating conditions, original frequency, adjusted frequency, and expected noise reduction effect. The control table adopts a hierarchical structure, categorized by factors such as load level, speed range, and environmental conditions. A real-time query mechanism for the control table is established to quickly retrieve the corresponding frequency adjustment parameters based on current operating conditions.

[0045] In step S140, spectrum analysis is performed on the enhanced control domain to identify high-energy frequency bands, frequency shifting operations are performed on the high-energy frequency bands to generate safe frequency band signals, energy aggregation processing is performed on the safe frequency band signals to form a frequency shift control sequence, and a multi-band management strategy is constructed based on the fusion of the frequency shift control sequence and the noise reduction control table.

[0046] Specifically, spectrum analysis of the enhanced control domain is performed to identify high-energy frequency bands. The acquired enhanced control domain data is input into a high-resolution spectrum analysis system, and a multi-window spectrum estimation method is used to improve the accuracy and stability of the spectrum analysis. The Thompson multi-window method is used to estimate the power spectral density, selecting an appropriate number of orthogonal window functions to balance frequency resolution and variance reduction. The energy density distribution P(f) = |F(f)|² / T is calculated for each frequency point within the enhanced control domain, where P(f) is the power spectral density at frequency f, F(f) is the Fourier transform of the enhanced control domain signal, and T is the observation time. High-energy frequency bands are identified using an energy density threshold method. The energy threshold is set to a certain percentage of the total energy, and continuous frequency intervals exceeding the threshold are marked as high-energy frequency bands. A sliding average filter is used to smooth the power spectrum curve to eliminate the influence of random fluctuations on frequency band identification. A peak detection algorithm is used to locate the energy peak in the power spectrum. The peak is then expanded from the peak to the sides until the energy density drops to half of the peak value, thus determining the boundaries of the high-energy frequency band. Establish a frequency band characteristic parameter library, recording key parameters such as center frequency, bandwidth, peak power, and integrated energy for each high-energy frequency band. Use time-frequency analysis to observe the time-varying characteristics of high-energy frequency bands and identify stable and transient frequency bands. Prioritize identified high-energy frequency bands, with those with high energy density and narrow bandwidth receiving higher processing priority. Establish a classification system for high-energy frequency bands, classifying them into fundamental-related, harmonic-related, resonance-related, and interference-related categories based on their physical origin.

[0047] In some embodiments, the frequency shifting operation performed on the high-energy frequency band to generate a safe frequency band signal includes: dividing the spectrum window in the high-energy frequency band to generate a spectrum window set; performing frequency shift path planning based on the spectrum window set to obtain a target frequency band; performing carrier modulation processing on the target frequency band to generate a shifted carrier; and using the shifted carrier to perform energy transfer processing to generate a safe frequency band signal.

[0048] Spectral windowing is performed within the high-energy frequency band to generate a set of spectral windows. The identified high-energy frequency band is processed, and the spectrum within the high-energy frequency band is subdivided. The principle of windowing is to maintain relatively uniform characteristics of the high-energy frequency band within each window, avoiding grouping significantly different frequency components into the same window. Energy gradient analysis is used to determine the window boundaries within the high-energy frequency band. The first-order derivative of the spectrum within the high-energy frequency band is calculated, ∇P(f) = dP(f) / df, where ∇P(f) is the power spectrum gradient at frequency f within the high-energy frequency band, P(f) is the corresponding power spectrum density, and d / df represents the differential operation with respect to frequency. Window demarcation points are set at locations within the high-energy frequency band where the gradient changes dramatically, ensuring smooth spectral changes within the window. An adaptive window size mechanism is established for the high-energy frequency band, dynamically adjusting the window size based on the complexity of the spectrum within the high-energy frequency band, using smaller windows for complex bands and larger windows for simple bands. A window overlap strategy is designed within the high-energy frequency band to maintain a certain overlap between adjacent windows to prevent boundary effects from affecting the processing results. Calculate the characteristic parameters of each window within the high-energy frequency band, including center frequency, effective bandwidth, integrated energy, and peak position. Establish an identification system for high-energy frequency band windows, assigning each window a unique number and attribute label. Use morphological processing to eliminate overly small windows within the high-energy frequency band and merge them into adjacent larger windows. Smooth the window boundaries within the high-energy frequency band, and use a transition function to achieve smooth connections between windows. The resulting spectrum window set contains complete window parameter information, providing a structured processing unit for frequency shift path planning.

[0049] Frequency shift paths are planned based on a set of spectrum windows to obtain the target frequency band. The generated set of spectrum windows is used as input, and a corresponding frequency shift path is designed for each spectrum window in the set. Path planning requires comprehensive consideration of multiple factors, including the availability of the target frequency band, the shift distance, and the energy capacity. A frequency space occupancy map is created, marking occupied and available frequency regions to provide constraints for path planning. The windows in the set are prioritized, and the optimal path is assigned to important windows. A graph search algorithm is used to find the optimal path from the source frequency band to the target frequency band. The path cost function incorporates factors such as the shift distance, the congestion level of the target area, and the shift complexity. An obstacle avoidance mechanism is designed to prevent the shift path from conflicting with occupied frequency bands. The path planning cost function is calculated as C = w1 × D + w2 × O + w3 × I, where C is the total cost, D is the shift distance, O is the target area occupancy, I is the implementation complexity, and w1, w2, and w3 are weight coefficients. Dynamic programming is used to solve the joint path planning problem for multiple windows, ensuring a global optimal solution. Establish a path conflict detection mechanism to identify intersections and overlaps between different window migration paths. Design a path adjustment strategy to automatically adjust the target frequency bands of some windows when conflicts occur. Determine the corresponding target frequency band for each window in the spectrum window set and establish a complete migration mapping relationship.

[0050] Carrier modulation is performed on the target frequency band to generate a shifted carrier. The goal of carrier modulation is to generate a carrier signal that accurately shifts the energy in the source frequency band to the target frequency band. The carrier frequency, f_carrier = f_target - f_source, is calculated based on the shift distance, where f_carrier is the carrier frequency, f_target is the center frequency of the target frequency band, and f_source is the center frequency of the source frequency band. The amplitude envelope of the carrier signal is designed, using a Gaussian envelope function to ensure frequency domain compactness. The carrier signal is represented as C(t) = A(t) × cos(2πf_carrier × t + φ(t)), where C(t) is the carrier signal, A(t) is the time-varying amplitude envelope, and φ(t) is the time-varying phase. Digital signal processing techniques are used to generate an accurate carrier signal, and direct digital frequency synthesizers are used to ensure frequency accuracy. A carrier parameter optimization algorithm is developed to adjust the carrier amplitude, frequency, and phase parameters based on feedback from the shifting effect. A multi-carrier coordination mechanism is designed to ensure orthogonality and interference-free operation between carriers when multiple carriers need to operate simultaneously. Establish a carrier quality monitoring system to monitor the frequency stability and amplitude consistency of carrier signals in real time. Use carrier pre-distortion technology to compensate for nonlinear distortion during the modulation process. Generate a carrier signal library to store carrier parameter templates corresponding to different migration requirements.

[0051] Energy transfer is performed using shifted carriers to generate a safe frequency band signal. The generated shifted carrier is modulated with the source frequency band signal to transfer energy from the high-energy frequency band to the safe frequency band. Energy transfer employs balanced modulation technology, achieving spectrum shifting through multiplication: M(t) = S(t) × C(t), where M(t) is the modulated signal, S(t) is the source frequency band signal, and C(t) is the shifted carrier. The modulated signal contains two spectral components: the upper sideband and the lower sideband. A filter is used to select the desired sideband signal. A sideband selection filter is designed, and a notch filter is used to suppress unwanted sidebands and carrier leakage. An energy transfer efficiency monitoring mechanism is established to calculate the energy ratio before and after the transfer to ensure the integrity of the energy transfer. Amplitude and phase compensation techniques are used to correct signal distortion during the transfer process. A multi-stage transfer scheme is designed. For long transfer distances, multiple smaller transfers are performed to achieve the final goal. A transfer quality assessment system is established, comprehensively considering metrics such as spectral purity, phase continuity, and amplitude fidelity. The generated safe frequency band signal is post-processed, including filtering, amplification, limiting and other operations.

[0052] Energy aggregation processing is performed on the safe frequency band signal to form a frequency shift control sequence. The goal of energy aggregation processing is to effectively integrate the energy dispersed across different safe frequency bands to form a sequence signal that is easy for the control system to process. Wavelet packet decomposition is used to perform multi-scale analysis on the safe frequency band signal, decomposing it into subbands with different frequency and time resolutions. The energy distribution of each subband is calculated as E_k=∫|W_k(t)|²dt, where E_k is the energy of the kth subband, W_k(t) is the wavelet coefficient of the kth subband, and the integral represents the integration over time. The importance weight of each frequency band is determined by calculating the energy centroid. The energy centroid frequency is f_cg=∫f×P(f)df / ∫P(f)df, where f_cg is the centroid frequency and P(f) is the power spectral density. An energy aggregation matrix is ​​established, with the matrix elements representing the energy correlation between different frequency bands. Principal component analysis is used to reduce the dimensionality of the energy aggregation matrix and extract the main energy aggregation patterns. A time-domain window function is designed to sample the aggregated energy over time, generating discrete control sequence points. The sampling frequency of the control sequence is determined based on the control system's response speed, ensuring that control instructions can promptly respond to system state changes. The generated frequency-shifted control sequence is smoothed to eliminate sudden changes and glitches, improving control stability. Coding rules for the control sequence are established to map continuous sequence values ​​into discrete control instructions.

[0053] A multi-band management strategy is constructed based on the fusion of frequency shift control sequences and noise reduction control tables. The fusion process requires coordinating frequency shift control and noise reduction control objectives to achieve effective frequency band management while ensuring effective noise reduction. A fusion weight allocation mechanism is established to prioritize frequency shift control and noise reduction control based on current operating conditions. A fusion algorithm is designed: F(t) = α × FC(t) + β × NC(t), where F(t) represents the fused management strategy, FC(t) represents the frequency shift control sequence, and NC(t) represents the noise reduction control command. α and β represent dynamic weight coefficients, satisfying the equation α + β = 1. Weight coefficients are determined based on real-time performance evaluation. β is increased when noise levels are high, and α is increased when frequency band conflicts are severe. A multi-band coordination mechanism is established to prevent conflicting control commands from different bands. A priority scheduling algorithm is used to handle concurrent control requests from multiple bands, with higher-priority control commands being executed first. A band switching strategy is designed to ensure smooth transition of management strategies under different operating conditions. An exception handling mechanism is established to automatically activate a backup management strategy when an exception occurs in a band. During the strategy execution process, the status changes of each frequency band are monitored in real time, and the management parameters are dynamically adjusted based on the feedback information.

[0054] Step S150 : decomposing the multi-band management strategy into parallel control flows, performing time-sequential staggering on the parallel control flows to generate a staggered control sequence, and generating an energy-saving control matrix based on the staggered control sequence.

[0055] Specifically, the multi-band management strategy is decomposed into parallel control flows. Each parallel control flow corresponds to a specific frequency band range and control task. Parallel processing improves system response speed and control accuracy. A functional decomposition approach is used to break down complex management strategies into several relatively simple control subtasks. A control flow classification system is established, classifying control flows into fundamental control flow, harmonic control flow, resonance suppression control flow, and interference cancellation control flow based on the physical characteristics of the frequency band. Each control flow includes independent input interfaces, processing logic, and output interfaces to ensure decoupling and independence between control flows. A control flow priority mechanism is designed to assign priorities to different control flows based on the importance and urgency of the frequency band. A communication protocol is established between control flows to support necessary data exchange and coordination. A pipeline architecture is used to organize parallel control flows, with each control flow running in an independent processing unit to avoid mutual interference. A load balancing algorithm is designed to ensure a relatively even computational load across control flows, preventing any one control flow from becoming a system bottleneck. A control flow status monitoring mechanism is established to track the operating status and processing progress of each control flow in real time. Design a fault-tolerant mechanism so that when a control flow fails, other control flows can continue to work normally. Record detailed information about the control flow decomposition, including the number of control flows, allocation strategy, interface definition, etc.

[0056] In some embodiments, the time-staggered arrangement of the parallel control flows to generate an interleaved control sequence includes: using the parallel control flows to perform switching moment analysis to establish a phase shift reference; using the phase shift reference to perform delay configuration processing to determine the phase shift moment; and using the phase shift moment to construct the interleaved control sequence.

[0057] Parallel control flows are used to analyze switching moments and establish a phase shift benchmark. The switching moments of each parallel control flow are analyzed, and a phase shift benchmark for the system is established through statistical analysis. Switching moment analysis identifies the temporal distribution characteristics of the switching actions of each control flow, providing a basis for subsequent phase shift design. The switching moment sequence of each control flow is extracted, and the occurrence and duration of the switching actions are recorded. Statistical characteristics of the switching moments are calculated, including parameters such as the mean interval, variance, skewness, and kurtosis. The periodic characteristics of the switching moments are analyzed, and Fourier analysis is used to identify the main periodic components. A probability distribution model for the switching moments is established, and the probability density function of the switching moment distribution is fitted using kernel density estimation. The correlation matrix R_ij = corr(T_i, T_j) of the switching moments of each control flow is calculated, where R_ij is the correlation coefficient between the switching moments of the i-th and j-th control flows, T_i and T_j are the corresponding switching moment sequences, and corr represents the correlation coefficient calculation function. Highly correlated control flow pairs are identified; these control flows require phase shifting to prevent simultaneous switching. Establish a phase shift reference frequency f_ref. This reference frequency is typically the system's fundamental frequency or an integer multiple thereof. Calculate the phase angle of each control flow relative to the reference frequency: φ_i = 2π × (t_i mod T_ref) / T_ref, where φ_i is the phase angle of the i-th control flow, t_i is the switching instant, T_ref is the reference period, and mod represents the modulo operation. Create a phase distribution plot to visualize the distribution of each control flow in phase space.

[0058] A phase-shift benchmark is used to perform delay configuration to determine the phase-shift timing. The goal of delay configuration is to introduce appropriate time delays to offset the timing of control flows that were originally in phase or close to each other, thereby reducing the peak power consumption caused by simultaneous switching. Using the reference frequency f_ref in the phase-shift benchmark as a synchronization reference, the phase deviation of each control flow relative to the benchmark is calculated. A delay optimization objective function is established, with the optimization goals of minimizing switching losses and current ripple. Based on the ideal phase distribution pattern set in the phase-shift benchmark, a delay allocation strategy is designed, employing the uniform distribution principle to evenly distribute each control flow within a benchmark cycle. Using the phase interval parameter in the phase-shift benchmark, the ideal phase interval Δφ_ideal = 2π / N is calculated, where N is the number of parallel control flows. Based on the phase-shift benchmark, each control flow is assigned a target phase φ_target_i = φ_base + i × Δφ_ideal, where φ_base is the base phase in the phase-shift benchmark and i is the control flow number. Calculate the required delay τ_i = (φ_target_i - φ_current_i) × T_ref / (2π), where τ_i is the delay of the i-th control flow, φ_current_i is the current phase, φ_target_i is the target phase determined based on the phase shift benchmark, and T_ref is the benchmark period. Establish delay constraints to ensure that the delay is within a reasonable range and avoid excessive delays that affect system response.

[0059] Use staggered timings to construct interleaved control sequences. The construction of interleaved control sequences requires comprehensive consideration of the priority, execution duration, and resource requirements of each control flow. A timeline model is established to divide a control cycle into several time units, each corresponding to a possible control action. The start time of each control flow is marked on the timeline based on the staggered timings. A control event scheduling algorithm is designed to ensure that no conflicting control actions are executed simultaneously at the same time. A coding format for the control sequence is established, using binary encoding to represent the state of each control flow at each moment. A control sequence matrix S(t,i) is generated, where S(t,i) represents the state of control flow i at time t, with 0 indicating off and 1 indicating on. Performance metrics for the interleaved control sequence are calculated, including switching frequency dispersion, peak power consumption reduction rate, and current ripple suppression. A cyclic detection system is implemented to ensure that the generated sequence exhibits good periodicity. A compressed storage method for the sequence is designed to reduce storage space requirements through pattern recognition. A real-time sequence generation mechanism is established to dynamically generate subsequent control sequences based on the current state. Record the detailed parameters of the interleaving control sequence, including sequence length, interleaving mode, performance indicators and other information.

[0060] In some embodiments, generating an energy-saving control matrix based on the staggered control sequence includes: identifying a switching loss concentration area through the staggered control sequence; performing energy distribution analysis on the switching loss concentration area to form a loss distribution graph; determining an optimization path based on the loss distribution graph; and forming an energy-saving control matrix through the optimization path.

[0061] Identify concentrated switching loss areas using staggered control sequences. The generated staggered control sequence is used as analysis input, and concentrated switching activity characteristics are identified from the sequence's temporal distribution pattern. Switching time data is extracted from the staggered control sequence, and the switching density function D(t) = N_switch(t) / Δt is calculated, where D(t) is the switching density at time t, and N_switch(t) is the number of switches within the time window Δt from the staggered control sequence. Based on the temporal characteristics of the staggered control sequence, a sliding window technique is used to analyze the time-varying characteristics of the switching density, with the window length determined based on the periodic characteristics of the staggered control sequence. A switching density threshold is set, and time periods exceeding the threshold are marked as concentrated switching loss areas. By analyzing the time intervals between adjacent control events in the staggered control sequence, cluster analysis is used to merge adjacent high-density time periods into continuous concentrated areas. Based on the execution pattern of the staggered control sequence, the causes of concentrated switching loss areas are analyzed, and the control strategies and sequence arrangements that lead to concentrated switching activity are identified. A severity assessment index for concentrated areas is established, taking into account factors such as switching frequency, power involved, and duration. Calculate the energy loss in each concentrated area: E_loss = ∑(V_switch × I_switch × t_switch), where V_switch is the switch voltage, I_switch is the switch current, and t_switch is the switching time. Prioritize the identified concentrated switching loss areas, prioritizing those with the highest losses and the most severe impact.

[0062] Energy distribution analysis is performed in the switching loss concentration area to generate a loss distribution map. A multidimensional loss analysis framework is established to analyze the loss distribution characteristics from the time, space, and frequency dimensions. The loss contribution of each control flow within the switching loss concentration area is calculated as P_loss_i = P_switch_i + P_conduction_i + P_blocking_i, where P_loss_i is the total loss of the i-th control flow, P_switch_i is the switching loss, P_conduction_i is the conduction loss, and P_blocking_i is the blocking loss. A heat map visualization technique is used to generate a loss distribution map, with time on the horizontal axis and the control flow number on the vertical axis. The color depth indicates the loss intensity. An interpolation algorithm is used to fill in the loss values ​​between discrete measurement points to generate a continuous loss distribution function. The statistical characteristics of the loss distribution are analyzed, and statistical quantities such as the mean, standard deviation, maximum, and minimum values ​​are calculated. Hot spots in the loss distribution are identified, which often correspond to weak links in the system. A loss gradient analysis is established to calculate the rate of change of loss in time and space. The principal component analysis method is used to extract the main patterns of loss distribution and identify the key factors affecting the loss distribution.

[0063] Determine the optimization path based on the loss distribution map. Based on the generated loss distribution map, design a systematic optimization strategy to determine the specific path and method for reducing switching losses. The design of the optimization path requires balancing loss reduction effectiveness with implementation complexity, selecting the most cost-effective optimization solution. Establish an optimization objective function with the primary goal of minimizing total loss while ensuring that system performance is not significantly impacted. Identify areas of potential optimization within the loss distribution map, focusing on areas with high loss density and large optimization potential. Design a multi-level optimization strategy, including timing optimization, parameter optimization, and topology optimization. Calculate the loss reduction potential of various optimization solutions: ΔP_reduction = P_original - P_optimized, where ΔP_reduction is the loss reduction, P_original is the original loss, and P_optimized is the optimized loss. Establish a feasibility evaluation system for the optimization solutions, comprehensively considering technical feasibility, economic feasibility, and implementation difficulty. Use a multi-objective optimization algorithm to find a Pareto-optimal solution set, balancing loss reduction with other performance indicators. Design a phased implementation plan for the optimization path, prioritizing optimization measures with rapid results and minimal risks. Establish an optimization effect evaluation method and verify the effectiveness of the optimization solutions through simulation and experiments. Record the detailed plan of the optimization path, including optimization steps, expected results, implementation plan and other information.

[0064] An energy-saving control matrix is ​​formed through the optimization path. Based on the different hierarchical strategies within the optimization path, including timing optimization, parameter optimization, and topology optimization, the matrix's dimensions are defined. The row dimension represents the implementation phase or operating conditions defined within the optimization path, and the column dimension represents the control parameters or control flow numbers involved in the optimization path. Based on the calculated loss reduction potential ΔP_reduction for each optimization solution within the optimization path, a matrix element selection rule is designed, with each element corresponding to the optimal value of a specific parameter at a specific phase. The phased implementation plan within the optimization path is converted into a matrix query mechanism, using a table lookup method for fast parameter query, allowing for rapid determination of the optimal control parameters based on the current implementation phase. Based on the continuity characteristic of the Pareto optimal solution set within the optimization path, a matrix interpolation mechanism is established to handle the unclear transition states between implementation phases. Based on the optimization effect evaluation method within the optimization path, an adaptive matrix update algorithm is designed to continuously optimize the matrix parameters based on actual operating results. Considering the implementation complexity control requirements within the optimization path, a compressed matrix storage format is established to reduce memory usage and access latency. Based on the multi-level parallel optimization strategy within the optimization path, a parallel matrix access mechanism is designed to support simultaneous parameter query by multiple control flows.

[0065] Step S160 , generating a PWM control signal based on the energy-saving control matrix, performing edge reconstruction on the PWM control signal to obtain a softened edge waveform, and outputting the softened edge waveform to the power device to complete intelligent control of the motor driver.

[0066] Specifically, PWM control signals are generated based on the energy-saving control matrix. The PWM signal generation process requires converting the control parameters in the matrix into specific pulse widths and switching timings. A PWM parameter mapping relationship is established, mapping the values ​​in the energy-saving control matrix to the duty cycle, frequency, and phase parameters of the PWM signal. The PWM duty cycle D = T_on / T_period is calculated, where D is the duty cycle, T_on is the high-level duration, and T_period is the PWM period. The PWM parameters are dynamically adjusted based on the motor's torque requirements and energy-saving optimization goals. Space vector pulse width modulation (SVPWM) technology is used to improve voltage utilization and reduce harmonic content. Symmetry constraints are established for the three-phase PWM signal to ensure balanced operation of the three-phase system. An adaptive PWM carrier frequency adjustment mechanism is designed to optimize the carrier frequency based on motor speed and load conditions. A PWM signal dithering technique is developed to disperse the energy concentration of the switching frequency through small random perturbations. Multi-level PWM technology is used to reduce voltage step amplitude and electromagnetic interference. A PWM signal synchronization mechanism is designed to ensure coordinated operation of multiple power devices. A PWM waveform quality monitoring system is established to monitor waveform symmetry, linearity, and stability in real time. The generated PWM control signal contains precise timing and amplitude information, providing instructions for precise control of power devices.

[0067] In some embodiments, the edge reconstruction of the PWM control signal to obtain a softened edge waveform includes: performing time domain expansion on the PWM control signal to obtain a transient voltage change rate; performing resonant peak detection on the transient voltage change rate to identify an EMI excitation point; constructing an anti-resonance injection sequence based on the EMI excitation point; and using the anti-resonance injection sequence to perform compensation modulation on the transient voltage change rate to obtain a softened edge waveform.

[0068] The PWM control signal is expanded in the time domain to obtain the transient voltage rate of change. The transient characteristics of PWM signals are the primary cause of electromagnetic interference and switching losses. Time domain expansion analysis can quantify these transient processes. High-precision time domain sampling technology is used to digitize the signal at a sampling rate significantly higher than the PWM carrier frequency. A time domain analysis window is established, focusing on the rising and falling edges of the PWM signal. The window length is determined by the edge duration. The transient voltage rate of change dV / dt = ΔV / Δt is calculated, where dV / dt is the voltage rate of change, ΔV is the voltage change, and Δt is the time interval. A numerical differentiation algorithm is used to calculate the voltage rate of change at discrete time points, and the central difference method is used to improve calculation accuracy. The time-varying characteristics of the transient voltage rate of change are analyzed to identify the peak location and amplitude of the rate of change. A statistical model of the transient characteristics is established to calculate statistical parameters such as the mean, standard deviation, maximum, and minimum values ​​of the rate of change. Wavelet transform technology is used to analyze the multi-scale characteristics of the transient process and identify transient components at different time scales. The spectrum characteristics of transient voltage change rate are established, and its frequency domain distribution is analyzed through Fourier transform.

[0069] Resonant peak detection is performed on the transient voltage rate of change to identify EMI excitation points. EMI excitation points typically correspond to extreme values ​​or sudden changes in the rate of change during the transient process; these points are the primary sources of electromagnetic energy radiation. Criteria for resonant peak detection are established, along with detection threshold and window parameters. A peak detection algorithm is used to locate local maxima in the transient voltage rate of change curve. The peak significance index S = (P_peak - P_background) / σ_noise is calculated, where S is the significance index, P_peak is the peak amplitude, P_background is the background level, and σ_noise is the noise standard deviation. Peaks whose significance index exceeds the set threshold are selected as candidate EMI excitation points. The frequency domain characteristics of the peak points are analyzed, and the frequency components corresponding to the peaks are analyzed using a short-time Fourier transform. An EMI excitation intensity assessment model is developed, comprehensively considering factors such as peak amplitude, frequency range, and duration. The spatial distribution characteristics of EMI excitation points are calculated, and the density of excitation points along the time axis is analyzed. A clustering algorithm is used to group similar EMI excitation points to identify concentrated areas of EMI excitation. Establish a priority ranking of EMI excitation points, giving priority to excitation points with high excitation intensity and wide impact range.

[0070] An anti-resonance injection sequence is constructed based on the EMI excitation point. The principle of anti-resonance injection is to inject a compensation signal with opposite phase and equal amplitude to the original signal at the EMI excitation point, achieving active interference suppression. The spectral characteristics of each EMI excitation point are analyzed to determine the frequency components and phase information to be suppressed. A generation model for the anti-resonance signal is established, and an anti-phase resonant signal corresponding to the EMI excitation frequency is designed. The amplitude of the anti-resonance signal is calculated as A_anti = -k × A_EMI, where A_anti is the anti-resonance signal amplitude, A_EMI is the amplitude at the EMI excitation point, and k is the compensation coefficient, with the negative sign indicating phase reversal. The phase of the anti-resonance signal is designed as φ_anti = φ_EMI + π, where φ_anti is the anti-resonance signal phase, φ_EMI is the phase at the EMI excitation point, and π represents a 180-degree phase difference. A multi-frequency anti-resonance mechanism is established to simultaneously address interference suppression at multiple EMI excitation points. An adaptive algorithm is used to dynamically adjust the anti-resonance parameters, optimizing the compensation strength in real time based on the suppression effect. Timing control for anti-resonance injection is designed to ensure that the compensation signal is injected into the system at the correct time. Establish an encoding format for anti-resonance injection sequences to support efficient storage and real-time recall. Calculate the power consumption cost of anti-resonance injection to ensure that the injection process does not significantly increase system losses. Record detailed parameters of the anti-resonance injection sequence, including key data such as injection time, signal amplitude, and phase information.

[0071] An anti-resonant injection sequence is used to compensate for the transient voltage rate of change, resulting in a soft-edged waveform. The goal of this compensation modulation is to significantly improve the transient characteristics and electromagnetic compatibility of the PWM signal while maintaining its basic functionality. A mathematical model for compensation modulation is established: V_soft(t)=V_original(t)+V_anti(t), where V_soft(t) is the softened signal, V_original(t) is the original PWM signal, and V_anti(t) is the anti-resonant injection signal. A compensation intensity control algorithm is designed to dynamically adjust the compensation parameters based on EMI suppression requirements and system performance requirements. Digital signal processing techniques are used to implement real-time compensation modulation, with a high-speed digital signal processor handling the compensation calculations. A compensation monitoring mechanism is established to evaluate the impact of the soft-edged signal on signal quality in real time. An edge smoothness evaluation metric is designed to quantify the effectiveness of the soft-edged signal. Filtering techniques are used to further smooth the compensated signal and eliminate high-frequency noise that may be introduced during the compensation process. A quality control system for the soft-edged waveform is established to ensure that the generated waveform meets the requirements of power device drivers. An optimization algorithm for the soft-edged parameters is designed, and a multi-objective optimization approach is used to find the optimal balance between EMI suppression and switching performance. A standardized format for softened waveforms is established to support compatibility with different types of power devices. The resulting softened edge waveforms have smooth transient characteristics, reducing electromagnetic interference levels while maintaining good switching control performance.

[0072] The softening edge waveform is output to the power device to complete the intelligent control of the motor driver. The output process needs to consider the characteristics of the drive circuit and the response characteristics of the power device to ensure the effective transmission of the control signal. An impedance matching network is established for the output interface to ensure the integrity and stability of the signal transmission process. An amplification circuit for the drive signal is designed to amplify the control signal to a level sufficient to drive the power device. An optical isolation mechanism is established to ensure electrical isolation and safety between the control circuit and the power circuit. A protection circuit for the power device is designed, including overcurrent protection, overvoltage protection, and overheat protection, etc. A feedback monitoring system for the drive signal is established to monitor the working state and response characteristics of the power device in real time. Dead-time control technology is used to avoid the shoot-through phenomenon of the upper and lower devices on the same bridge arm. A temperature compensation mechanism for the drive circuit is designed to ensure stable operation at different temperatures. A fault diagnosis and handling mechanism is established to automatically switch to a safe mode when an abnormal condition is detected. A soft start function is designed for the system to avoid start-up impact by gradually increasing the output voltage. Finally, the intelligent control of the motor driver is completed, realizing a complete control link from signal generation to power output, with the characteristics of high efficiency, low noise, and intelligence.

[0073] In order to perform the intelligent control method of a motor driver corresponding to the above-mentioned method embodiment, to realize the corresponding functions and technical effects. Referring to Figure 2 , Figure 2 The structure block diagram of the intelligent control system 200 of a motor driver provided by the embodiment of the present application is shown. For ease of illustration, only the parts related to the present embodiment are shown. The intelligent control system 200 of a motor driver provided by the embodiment of the present application comprises: The signal acquisition module 201 is configured to acquire multi-source abnormal signals in the motor operation process, wherein the multi-source abnormal signals include harmonic currents and sound pressure fluctuation data, and perform feature conversion on the multi-source abnormal signals to form a fault feature library. The energy recovery module 202 is configured to identify available fault modes based on the fault feature library, perform energy recovery evaluation on the available fault modes to determine energy collection points, perform harmonic phase shift processing on the energy collection points to generate auxiliary drive signals, and perform amplitude-frequency characteristic extraction on the auxiliary drive signals to determine an enhanced control domain. The acoustic optimization module 203 is configured to perform frequency spectrum analysis on the sound pressure fluctuation data to generate an acoustic feature spectrum, perform feature decomposition on the acoustic feature spectrum to obtain an efficiency indicator sound, map the efficiency indicator sound to a torque optimization parameter, perform electromagnetic excitation frequency adjustment based on the torque optimization parameter to establish a noise reduction control table. a frequency management module 204 configured to perform spectrum analysis on the enhanced control domain to identify high-energy frequency bands, perform frequency shifting on the high-energy frequency bands to generate safe frequency band signals, perform energy aggregation processing on the safe frequency band signals to form a frequency shift control sequence, and construct a multi-band management strategy based on the fusion of the frequency shift control sequence and the noise reduction control table; A timing control module 205 is configured to decompose the multi-band management strategy into parallel control flows, perform time-staggered arrangement on the parallel control flows to generate an interleaved control sequence, and generate an energy-saving control matrix based on the interleaved control sequence; The output shaping module 206 is used to generate a PWM control signal based on the energy-saving control matrix, perform edge reconstruction on the PWM control signal to obtain a softened edge waveform, and output the softened edge waveform to the power device to complete the intelligent control of the motor driver.

[0074] The intelligent control system 200 for a motor drive described above can implement the intelligent control method for a motor drive described in the method embodiment described above. The optional options in the method embodiment described above also apply to this embodiment and will not be described in detail here. The remaining contents of the embodiments of this application can be referred to the contents of the method embodiment described above and will not be described in detail in this embodiment.

[0075] The purpose of the above embodiments is to exemplify and deduce the technical solution of the present invention and to fully describe the technical solution, purpose and effect of the present invention. Its purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosed content of the present invention, and it does not limit the scope of protection of the present invention.

[0076] The above embodiments are not exhaustive and may include many other embodiments not listed above. Any replacements and improvements made without violating the concept of the present invention are within the scope of protection of the present invention.

Claims

1. An intelligent control method for a motor driver, characterized in that: include: Collecting multi-source abnormal signals during motor operation, the multi-source abnormal signals including harmonic current and sound pressure fluctuation data, and performing feature conversion on the multi-source abnormal signals to form a fault feature library; Identifying an exploitable fault mode based on the fault feature library, performing energy recovery evaluation on the exploitable fault mode to determine an energy collection point, performing harmonic phase shift processing on the energy collection point to generate an auxiliary drive signal, and extracting amplitude-frequency characteristics of the auxiliary drive signal to determine an enhanced control domain; Performing spectral analysis on the sound pressure fluctuation data to generate an acoustic characteristic spectrum, performing feature decomposition on the acoustic characteristic spectrum to obtain an efficiency indicator sound, mapping the efficiency indicator sound into a torque optimization parameter, and adjusting the electromagnetic excitation frequency based on the torque optimization parameter to establish a noise reduction control table; Performing spectrum analysis on the enhanced control domain to identify high-energy frequency bands, performing frequency shifting operations on the high-energy frequency bands to generate safe frequency band signals, performing energy aggregation processing on the safe frequency band signals to form a frequency shift control sequence, and constructing a multi-band management strategy based on the fusion of the frequency shift control sequence and the noise reduction control table; Decomposing the multi-band management strategy into parallel control flows, staggering the parallel control flows in time sequence to generate a staggered control sequence, and generating an energy-saving control matrix based on the staggered control sequence; A PWM control signal is generated based on the energy-saving control matrix, edge reconstruction is performed on the PWM control signal to obtain a softened edge waveform, and the softened edge waveform is output to a power device to complete intelligent control of the motor driver.

2. The intelligent control method for a motor driver according to claim 1, characterized in that: The performing feature conversion on the multi-source abnormal signals to form a fault feature library includes: constructing a current distortion map based on the harmonic current; Performing cycle-to-cycle difference calculation on the current distortion spectrum to identify the distortion growth rate; Constructing an acoustic-electrical correlation matrix based on the distortion growth rate and the sound pressure fluctuation data; A fault feature library is established based on the acoustic-electrical correlation matrix.

3. The intelligent control method of the motor driver according to claim 1, characterized in that: The performing energy recovery evaluation on the available fault mode to determine the energy collection point includes: extracting energy distribution of the harmonic current from the available fault mode; determining a resonance region based on the energy distribution; Performing energy density evaluation on the resonance region to generate a high-density collection area; Energy collection points are determined based on the high-density collection areas.

4. The intelligent control method of the motor driver according to claim 1, characterized in that: The performing feature decomposition on the acoustic feature spectrum to obtain the efficiency indicator sound includes: Performing frequency component separation on the acoustic characteristic spectrum to extract efficiency-related components; Performing feature screening processing on the efficiency-related components to generate a dominant frequency set; Performing correlation analysis on the dominant frequency set and the harmonic current to obtain an efficiency characteristic value; An efficiency indication sound is extracted according to the efficiency feature value.

5. The intelligent control method of the motor driver according to claim 1, characterized in that: The performing a frequency shift operation on the high-energy frequency band to generate a safe frequency band signal includes: Performing spectrum window division in the high-energy frequency band to generate a spectrum window set; Performing frequency shift path planning based on the spectrum window set to obtain a target frequency band; Performing carrier modulation processing on the target frequency band to generate a shifted carrier; The shifted carrier is used to perform energy transfer processing to generate a safe frequency band signal.

6. The intelligent control method of the motor driver according to claim 1, characterized in that: The step of performing time-sequential staggering on the parallel control flows to generate a staggered control sequence includes: Using the parallel control flow to perform switching moment analysis to establish a phase shift benchmark; Determine the phase shift moment by performing delay configuration processing using the phase shift reference; The staggered phase moments are used to construct an interleaved control sequence.

7. The intelligent control method of a motor driver according to claim 1, characterized in that: The generating of an energy-saving control matrix based on the staggered control sequence includes: identifying a switching loss concentration area by using the staggered control sequence; Performing energy distribution analysis on the switching loss concentration area to form a loss distribution diagram; determining an optimization path according to the loss distribution map; An energy-saving control matrix is ​​formed through the optimization path.

8. The intelligent control method for a motor driver according to claim 1, characterized in that: The edge reconstruction of the PWM control signal to obtain a softened edge waveform includes: Expanding the PWM control signal in the time domain to obtain a transient voltage change rate; Performing resonance peak detection on the transient voltage change rate to identify EMI excitation points; Constructing an anti-resonance injection sequence based on the EMI excitation point; The anti-resonance injection sequence is used to perform compensation modulation on the transient voltage change rate to obtain a softened edge waveform.

9. The intelligent control method for a motor driver according to claim 3, characterized in that: The determining of the resonance region based on the energy distribution includes: Determining search accuracy by evaluating positioning complexity based on the energy distribution, wherein the positioning complexity includes the number of energy peaks, distribution dispersion, and harmonic order; Setting area division parameters according to the search accuracy; The energy distribution is spatially mapped using the region division parameters to generate resonance regions.

10. An intelligent control system for a motor driver, characterized in that: include: A signal acquisition module is used to collect multi-source abnormal signals during the operation of the motor, the multi-source abnormal signals including harmonic current and sound pressure fluctuation data, and perform feature conversion on the multi-source abnormal signals to form a fault feature library; an energy recovery module, configured to identify exploitable fault modes based on the fault signature library, perform energy recovery evaluation on the exploitable fault modes to determine energy collection points, perform harmonic phase shift processing on the energy collection points to generate auxiliary drive signals, and extract amplitude-frequency characteristics of the auxiliary drive signals to determine an enhanced control domain; an acoustic optimization module, configured to perform spectral analysis on the sound pressure fluctuation data to generate an acoustic characteristic spectrum, perform feature decomposition on the acoustic characteristic spectrum to obtain an efficiency indicator sound, map the efficiency indicator sound into a torque optimization parameter, and adjust the electromagnetic excitation frequency based on the torque optimization parameter to establish a noise reduction control table; a frequency management module configured to perform spectrum analysis on the enhanced control domain to identify high-energy frequency bands, perform frequency shifting on the high-energy frequency bands to generate safe frequency band signals, perform energy aggregation processing on the safe frequency band signals to form a frequency shift control sequence, and construct a multi-band management strategy based on the fusion of the frequency shift control sequence and the noise reduction control table; a timing control module, configured to decompose the multi-band management strategy into parallel control flows, perform time-staggered arrangement on the parallel control flows to generate an interleaved control sequence, and generate an energy-saving control matrix based on the interleaved control sequence; An output shaping module is used to generate a PWM control signal based on the energy-saving control matrix, perform edge reconstruction on the PWM control signal to obtain a softened edge waveform, and output the softened edge waveform to the power device to complete the intelligent control of the motor driver.

Images