Motor drive cooperative control method under high-frequency switching circuit

Abstract

The application provides a motor driving cooperative control method under a high-frequency switching circuit, comprising: acquiring a PWM signal duty cycle of the high-frequency switching circuit in a current operation period for driving a motor, and determining a conduction window width of the high-frequency switching circuit in the current operation period according to the signal duty cycle; collecting a feedback signal at the sampling point, filtering the feedback signal to extract a first stable current, and evaluating whether the first stable current is consistent with a second stable current obtained by the same process in a previous operation period; when the first stable current is consistent with the second stable current, it is determined that the sampling point falls into a current stable area, the first stable current is output as a target feedback signal to a closed-loop control module, and torque cooperative control of the motor driving under the high-frequency switching circuit is completed.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for coordinated control of motor drive under high-frequency switching circuit. Background Technology

[0002] Motor drive control using high-frequency switching circuits is a core component of modern industrial and new energy equipment. The accuracy of closed-loop control of motor torque directly determines the stability and energy efficiency of equipment operation. Obtaining accurate current feedback signals is the cornerstone of high-precision control. In this process, the controller needs to monitor the phase current flowing through the motor windings in real time, calculate the actual output torque based on these current values, and then adjust the pulse width modulation signal to track the target torque command.

[0003] The effectiveness of this closed-loop control mechanism depends entirely on the accuracy of current sampling. Any sampling deviation will be directly transmitted to the torque calculation result, causing inaccurate control response. However, conventional current sampling methods usually set the sampling time based on empirical parameters under specific operating conditions, lacking adaptability to dynamic changes in motor operating status. To avoid the severe ringing noise generated when the switching transistor is first turned on, the common practice is to directly extend the sampling trigger waiting time. At the moment the switching transistor is turned on, due to the interaction of parasitic inductance and parasitic capacitance in the circuit, a high-frequency oscillation signal will be superimposed on the current detection path. This oscillation may last from hundreds of nanoseconds to several microseconds, and its amplitude often far exceeds the actual current signal, causing serious measurement pollution. In order to ensure that the sampling point is far away from this noise pollution area, the traditional solution sets a fixed delay time window after the switching transistor is turned on, waiting for the ringing to be sufficiently reduced before starting the analog-to-digital converter for sampling. Although this approach avoids the initial noise, it ignores the squeezing effect of drastic changes in motor operating status on the entire sampling cycle, resulting in severe distortion of the feedback signal under complex operating conditions. This distortion stems from the contradiction between the selection of sampling time in a high-frequency switching environment and the dynamic operation of the motor.

[0004] As the motor speed gradually increases, the duty cycle of the pulse width modulation signal dynamically narrows, meaning the effective window period for the switching transistor to remain on is significantly shortened. Within this severely compressed, extremely short window, maintaining a longer delay to avoid preceding ringing noise will trigger a chain reaction. Specifically, it's like rushing to move supplies in front of a rapidly closing gate; delaying action to wait for the dust to settle results in the gate closing before starting to move, leading to mission failure. In motor control, excessive delays cause the sampling point to slip directly out of the current's stable phase and into a region where the current is not yet stable or has even begun to decline, introducing entirely new sampling biases and completely destroying the accuracy of torque closed-loop control.

[0005] Therefore, under the trend of the duty cycle dynamically narrowing with the rotational speed, accurately grasping the balance between reducing the ringing noise of the switch and the remaining width of the conduction window, dynamically identifying and ensuring that the sampling point always falls into the true current stable region, becomes the key issue to achieve real and reliable torque closed-loop feedback and stable control accuracy. Summary of the Invention

[0006] This invention provides a method for coordinated control of motor drive under high-frequency switching circuit, comprising:

[0007] Obtain the duty cycle of the PWM signal used by the high-frequency switching circuit to drive the motor, and determine the conduction window width of the high-frequency switching circuit in the current operating cycle based on the signal duty cycle;

[0008] Collect ringing noise signals within the width of the conduction window, evaluate whether the peak voltage of the ringing noise signals is higher than a preset voltage threshold, and determine that there is effective ringing noise within the conduction window when the peak voltage is higher than the preset voltage threshold. Then, extract the duration of the ringing noise signals to obtain the ringing duration.

[0009] Calculate the proportion of the ringing duration within the width of the conduction window. When the proportion is lower than a preset proportion threshold, it is determined that the ringing noise has subsided in the first part of the conduction window and the current in the remaining section of the window tends to be stable. The preset initial sampling time is delayed backward on the time axis by the ringing duration to obtain the sampling point.

[0010] When the time ratio is not lower than the preset ratio threshold, it is determined that the ringing noise occupies too much of the conduction window and there is not enough remaining stable section. The preset initial sampling time is delayed backward on the time axis by the width of the conduction window to obtain the sampling point.

[0011] A feedback signal is collected at the sampling point, the feedback signal is filtered to extract the first stable current, and the consistency between the first stable current and the second stable current obtained by the same process in the previous running cycle is evaluated.

[0012] When the sampling points converge, it is determined that they fall into the current stable region. The first stable current is then output as the target feedback signal to the closed-loop control module to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

[0013] Preferably, the step of obtaining the duty cycle of the PWM signal used by the high-frequency switching circuit to drive the motor, and determining the conduction window width of the high-frequency switching circuit in the current operating cycle based on the signal duty cycle, includes:

[0014] Obtain PWM signal duty cycle information, and determine the conduction window width within the current operating cycle based on the correspondence between the duty cycle information and the switching cycle;

[0015] The conduction window segment is divided based on the conduction window width, and a conduction window segment description is formed by combining the duty information and the relationship between the switching cycle.

[0016] Consistency processing is performed on the description of the conduction window segment, and the conduction window width result is obtained based on the conduction window width.

[0017] Preferably, the step of acquiring ringing noise signals within the width of the conduction window, evaluating whether the peak voltage of the ringing noise signals is higher than a preset voltage threshold, and determining that there is valid ringing noise within the conduction window when the peak voltage is higher than the preset voltage threshold, then extracting the duration of the ringing noise signals to obtain the ringing duration includes:

[0018] Obtain the window start point and window end point corresponding to the width of the conduction window, collect the transient voltage waveform of the current detection path between the window start point and the window end point, and extract the ringing noise signal from the transient voltage waveform;

[0019] For the ringing noise signal, the voltage amplitude of each sampling point is read, and the largest amplitude value is selected from the voltage amplitude as the peak voltage. When the peak voltage is higher than the preset voltage threshold, an over-limit ringing mark is obtained.

[0020] The continuous over-limit segment formed by locating adjacent over-limit sampling points based on the over-limit ringing mark is determined by the starting and ending boundaries of the continuous over-limit segment, and the ringing duration is obtained.

[0021] Preferably, the calculation of the ringing duration as a proportion of the time within the conduction window width, and determining that the ringing noise has subsided in the first part of the conduction window and the current in the remaining section of the window tends to stabilize when the time proportion is lower than a preset proportion threshold, involves delaying the preset initial sampling time backward on the time axis using the ringing duration to obtain the sampling point, including:

[0022] Obtain the ringing duration and the conduction window width, and extract the duration of the ringing duration and the duration of the conduction window width according to the same time base to form a window occupancy record;

[0023] Based on the window occupancy record, the duration of the ringing is proportionally converted to the duration of the conduction window width to obtain the time ratio;

[0024] If the time ratio is lower than the preset ratio threshold, it is determined that the ringing noise subsides in the first part of the conduction window and the current in the remaining section of the window tends to be stable. The preset initial sampling time is then shifted back along the time axis by the duration of the ringing noise to obtain the sampling point.

[0025] Preferably, the step of determining that the ringing noise occupies too large a conduction window and the remaining stable segment is insufficient when the time ratio is not lower than a preset ratio threshold, and obtaining the sampling point by delaying the preset initial sampling time backward on the time axis with the conduction window width, includes:

[0026] Obtain the time ratio and a preset ratio threshold, compare their sizes under the same time base, and if the time ratio is not lower than the preset ratio threshold, then obtain a window over-occupancy flag;

[0027] Based on the window over-occupancy marker, it is determined that the ringing noise occupies too much of the conduction window, and that the remaining smooth section within the conduction window is insufficient.

[0028] Based on the determination that the remaining stable section is insufficient, the width of the conduction window is used as the shift time, so that the preset initial sampling time is delayed backward along the time axis to obtain the sampling point.

[0029] Preferably, the step of acquiring a feedback signal at the sampling point, filtering the feedback signal to extract a first stable current, and evaluating whether the first stable current is consistent with the second stable current obtained through the same process in the previous operating cycle includes:

[0030] The sampling point is obtained, and the current detection path is triggered at the sampling point to perform sampling. The feedback voltage waveform corresponding to the sampling point is collected, and the feedback voltage waveform is used as the feedback signal.

[0031] The feedback signal is low-pass filtered to remove high-frequency fluctuation components, resulting in a filtered feedback signal. The first stable current of the current operating cycle is then extracted from the filtered feedback signal.

[0032] The second stable current obtained from the previous running cycle using the same sampling and filtering process is acquired. The amplitude difference between the first stable current and the second stable current is compared. If the amplitude difference is lower than a preset consistency threshold, it is determined that the two are approaching consistency.

[0033] Preferably, the step of determining that the sampling point falls into the current stable region when they tend to be consistent, and outputting the first stable current as the target feedback signal to the closed-loop control module to complete the torque coordinated control of the motor drive under the high-frequency switching circuit includes:

[0034] Obtain the result of the first stable current and the second stable current converging. If the result is converging, then obtain the current stable region marker based on the sampling point.

[0035] The first stable current is read according to the current stable region mark, and the first stable current is used as the target feedback signal to obtain target feedback data with the current operating cycle identifier.

[0036] The target feedback data is used to output the target feedback signal to the closed-loop control module. The closed-loop control module calculates the pulse width modulation correction amount based on the deviation between the target feedback signal and the target torque command, and adjusts the switching timing of the high-frequency switching circuit based on the pulse width modulation correction amount to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

[0037] Preferably, the step of reading the voltage amplitude at each sampling point of the ringing noise signal, selecting the largest amplitude from the voltage amplitudes as the peak voltage, and obtaining an over-limit ringing flag when the peak voltage is higher than a preset voltage threshold includes:

[0038] The continuous over-limit segment formed by locating adjacent over-limit sampling points based on the over-limit ringing mark is determined by the starting and ending boundaries of the continuous over-limit segment, and the ringing duration is obtained.

[0039] Preferably, the step of performing low-pass filtering on the feedback signal to remove high-frequency fluctuation components in the feedback signal to obtain a filtered feedback signal, and extracting the first stable current of the current operating cycle from the filtered feedback signal, includes:

[0040] The second stable current obtained from the previous running cycle using the same sampling and filtering process is acquired. The amplitude difference between the first stable current and the second stable current is compared. If the amplitude difference is lower than a preset consistency threshold, it is determined that the two are approaching consistency.

[0041] Preferably, the step of reading the first stable current according to the current stability region marker and using the first stable current as the target feedback signal to obtain target feedback data with the current operating cycle identifier includes:

[0042] The target feedback data is used to output the target feedback signal to the closed-loop control module. The closed-loop control module calculates the pulse width modulation correction amount based on the deviation between the target feedback signal and the target torque command, and adjusts the switching timing of the high-frequency switching circuit based on the pulse width modulation correction amount to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

[0043] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0044] This invention discloses a motor drive collaborative control method under a high-frequency switching circuit. Addressing the core issue of the accuracy of sampling feedback signals to counteract ringing noise interference generated during motor drive, a dynamic adaptive sampling point optimization scheme is proposed. This method determines the conduction window width by acquiring the PWM signal duty cycle. Within the window, the ringing noise signal is acquired in real time, and its peak voltage and duration are evaluated. When the ringing noise duration accounts for a low proportion of the conduction window, it indicates that the noise has subsided in the early part of the window and the current tends to stabilize in the later part. In this case, delaying the sampling time by the ringing duration avoids the interference region. When the proportion is too high, it indicates that the stable section is insufficient, and the entire conduction window width is directly delayed to ensure that the sampling falls within the next stable cycle. The consistency between the first stable current obtained from the current sampling and the second stable current of the previous cycle is compared to verify whether the sampling point accurately falls within the current stable region. Finally, the verified stable current is used as the target feedback signal and output to the closed-loop control module, thereby effectively avoiding the pollution of the feedback signal by ringing noise and achieving precise torque collaborative control of the motor drive under a high-frequency switching circuit. Attached Figure Description

[0045] Figure 1 This is a flowchart of a motor drive coordinated control method under a high-frequency switching circuit according to the present invention.

[0046] Figure 2 This is a schematic diagram of a motor drive coordinated control method under a high-frequency switching circuit according to the present invention.

[0047] Figure 3 This is another schematic diagram of a motor drive coordinated control method under a high-frequency switching circuit according to the present invention. Detailed Implementation

[0048] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0049] like Figures 1-3 This embodiment of a motor drive coordinated control method under a high-frequency switching circuit may specifically include:

[0050] S101. Obtain the duty cycle of the PWM signal used by the high-frequency switching circuit to drive the motor in the current operating cycle, and determine the conduction window width of the high-frequency switching circuit in the current operating cycle based on the signal duty cycle.

[0051] The duty cycle information of the PWM signal is obtained. Within the current operating cycle, the conduction window width is determined based on the correspondence between the duty cycle information and the switching cycle. The conduction window is divided into segments based on the conduction window width, and a description of each conduction window segment is formed by combining the duty cycle information and the switching cycle. Consistency processing is performed on the conduction window segment descriptions, and the conduction window width result is obtained based on the conduction window width.

[0052] In one implementation, the PWM signal duty cycle information is used to characterize the time ratio between the duration of the high-level signal and the total duration of the cycle within a switching cycle. The switching cycle is formed by the high-frequency switching beat of the drive motor, and the conduction window width is mapped from this ratio. The duty cycle information is obtained by time sampling the high-low level switching points of the PWM waveform and aligning it with the switching cycle to form a unified time reference.

[0053] For example, the PWM signal obtains the duration interval between the rising and falling edges at the sampling time through edge detection, and this interval corresponds to the high-level duration. The ratio of the high-level duration to the switching period is used to describe the structural characteristics of the duty cycle information and serves as the basic data source for calculating the conduction window width.

[0054] In one possible implementation, the switching cycle is divided by the drive motor control clock signal, and each cycle boundary is determined by a cycle synchronization marker. The PWM duty cycle is truncated within the cycle boundaries and normalized and aligned according to the cycle boundaries, so that the duty cycles of different cycles are in a unified time reference system, thereby forming a basic time series that can be used to determine the window width. In a more refined implementation, the conduction window width is directly obtained by multiplying the PWM duty cycle by the switching cycle.

[0055] Specifically, the PWM duty cycle is denoted as D, and the switching period is denoted as T. sw Then the width of the conduction window T on =D×T swThe product result represents the length of the PWM high-level interval on the time axis, and is truncated under period boundary constraints to ensure consistency with the current operating cycle. In one embodiment, the conduction window width is further corrected during the formation process through an edge timestamp alignment operation within the cycle. Specifically, the edge timestamp alignment operation involves reading the rising edge of the PWM signal as the window start timestamp and the falling edge of the PWM signal as the window end timestamp; the difference between the two timestamps is the corrected conduction window width, thus forming a complete conduction interval expression. This conduction interval maintains structural consistency across different operating cycles and serves as the basis for subsequent segmentation. In another embodiment, the conduction window segment is composed of the interval range of the start and end timestamps on the time axis. The segment includes a high-level duration region and rising / falling edge boundary points, and the segment structure is formed by time calibration of the boundary points. The segment structure is used to characterize the stable conduction range of the PWM signal within a single cycle. In one implementation, the conduction window width is determined as a stable time interval within the current operating cycle and maintains a corresponding relationship with the PWM duty cycle, thereby completing the mapping relationship between the conduction window width and the PWM duty cycle, so that the conduction interval expression of the high-frequency switching circuit within the current operating cycle maintains a deterministic structure under a unified time reference.

[0056] S102. Collect the ringing noise signal within the width of the conduction window, evaluate whether the peak voltage of the ringing noise signal is higher than a preset voltage threshold, and when the peak voltage is higher than the preset voltage threshold, determine that there is effective ringing noise within the conduction window, and then extract the duration of the ringing noise signal to obtain the ringing duration.

[0057] The start and end points of the conduction window corresponding to its width are obtained. Transient voltage waveforms of the current detection path are acquired between the start and end points of the window, and ringing noise signals are extracted from these waveforms. The voltage amplitude of each sampling point is read from the ringing noise signal, and the largest amplitude is selected as the peak voltage. When the peak voltage exceeds a preset voltage threshold, an over-limit ringing marker is obtained. Based on the over-limit ringing markers, a continuous over-limit segment formed by adjacent over-limit sampling points is located. The width of the ringing segment is determined using the start and end boundaries of the continuous over-limit segment, and the ringing duration is obtained.

[0058] In one implementation, the conduction window width is determined by the PWM conduction interval within the current operating cycle. The window start point corresponds to the time position when the switch enters the conduction state, and the window end point corresponds to the time position when the switch exits the conduction state. The sampling unit reads the voltage change on the current detection path between the window start point and the window end point, forming a transient voltage waveform. The transient voltage waveform includes a high-frequency fluctuation portion at the moment the switch turns on and a relatively smooth portion at the end of the conduction window. A ringing noise signal is extracted from the transient voltage waveform. The extraction time starts at the moment t0 when the rising edge of the switch drive signal is detected, and ends at the moment t1 when the conduction window ends, where the time interval between t1 and t0 is equal to the preset conduction window width. The extracted ringing noise signal includes the complete high-frequency fluctuation portion starting from the moment of conduction and the subsequent smooth portion, ensuring that the ringing noise signal maintains the same time reference as the current operating cycle. The extraction process is achieved by a high-speed sampling circuit continuously recording the transient voltage value within the time interval t0 to t1.

[0059] For example, during the conduction of the motor drive bridge arm, a voltage fluctuation that decays over time will appear at the output of the current sensing resistor or current sensing amplifier. The sampling unit records the voltage amplitude of each sampling point according to a fixed sampling cycle, and the voltage amplitudes are arranged in chronological order of sampling time to form a sampling point sequence corresponding to the transient voltage waveform.

[0060] In one possible implementation, the peak voltage of the ringing noise signal is obtained by amplitude filtering of the sampling point sequence. Specifically, voltage amplitudes are read from all sampling points corresponding to the ringing noise signal, and the magnitudes of each voltage amplitude are compared. The voltage corresponding to the sampling point with the highest amplitude is determined as the peak voltage. If both positive and negative ringing exist, the sampling point with the largest deviation from the stable reference voltage is taken as the peak voltage point. The stable reference voltage is obtained as follows: at the end of the conduction window, an interval where the voltage change rate ΔV of N consecutive sampling points is less than 0.5% is identified as a smooth voltage interval, where N is between 8 and 16, and ΔV is the ratio of the voltage difference between adjacent sampling points to the voltage of the previous sampling point. Then, the arithmetic mean of the voltages of all sampling points within this smooth voltage interval is calculated as the stable reference voltage V. ref This processing method ensures that the peak voltage covers both unidirectional overshoot and high-amplitude fluctuations caused by reverse oscillation. Furthermore, a preset voltage threshold is used to distinguish between ringing noise signals and normal voltage fluctuations. The preset voltage threshold is determined based on statistical analysis of voltage fluctuations during the stable conduction phase of the current detection path. Specifically, after operating the motor under rated load for several calibration cycles, voltage samples are collected from the smoother area after the conduction window, and their standard deviation σ is calculated. v The preset voltage threshold is set to be equal to the stable reference voltage V. refAdding k times the standard deviation (k ranges from 3 to 5), the typical value is V. ref +0.2V is written to the threshold storage area during the controller initialization phase. The peak voltage is compared with the preset voltage threshold. If the peak voltage is higher than the preset voltage threshold, it is determined that there is effective ringing noise within the conduction window.

[0061] Preferably, when the peak voltage is higher than a preset voltage threshold, not only are the sampling points corresponding to the peak voltage recorded, but also the sampling points in the ringing noise signal that are higher than the preset voltage threshold are marked. The marked sampling points form over-limit ringing marks, while the sampling points that are not higher than the preset voltage threshold do not form over-limit ringing marks.

[0062] Specifically, a continuous over-limit segment consists of sampling points that are temporally adjacent and both carry the over-limit ringing marker. When there are no non-over-limit sampling points between two over-limit sampling points, they are grouped into the same continuous over-limit segment; when there are non-over-limit sampling points between adjacent over-limit sampling points, the previous continuous over-limit segment terminates, and the next over-limit sampling point forms a new continuous over-limit segment.

[0063] In one embodiment, if multiple consecutive over-limit segments exist within the ringing noise signal, it is first detected whether each segment contains a sampling point corresponding to the peak voltage. If only one consecutive over-limit segment contains a sampling point corresponding to the peak voltage, then that segment is directly selected as the ringing segment. If multiple consecutive over-limit segments all contain fluctuations exceeding a preset voltage threshold, then the temporal correlation of these segments needs to be further determined. When the interval between adjacent over-limit segments is less than 5 microseconds, these adjacent segments are merged into one ringing segment in chronological order, and the merged segment is taken as the final ringing segment. The starting boundary of the ringing segment is determined by the earliest over-limit sampling point within the ringing segment, and the ending boundary of the ringing segment is determined by the last over-limit sampling point within the ringing segment. The sampling unit reads the sampling time corresponding to the starting boundary and the sampling time corresponding to the ending boundary, and determines the ringing segment width based on the time span between the two. In one embodiment, the ringing segment width is output as the ringing duration. The ringing duration and the conduction window width use the same time unit and are stored in correspondence with the current operating cycle, so that the determination result of the existence of effective ringing noise within the conduction window and the duration result have a consistent time domain description.

[0064] S103. Calculate the proportion of the ringing duration within the width of the conduction window. When the proportion of the duration is lower than a preset proportion threshold, it is determined that the ringing noise has subsided in the first part of the conduction window and the current in the remaining section of the window tends to be stable. The preset initial sampling time is delayed backward on the time axis by the ringing duration to obtain the sampling point.

[0065] The ringing duration and the conduction window width are obtained. The duration of the ringing noise and the width of the conduction window are extracted using the same time reference to form a window occupancy record. Based on the window occupancy record, the duration of the ringing noise and the width of the conduction window are proportionally converted to obtain a time ratio. If the time ratio is lower than a preset threshold, it is determined that the ringing noise subsides in the first part of the conduction window and the current in the remaining section of the window tends to stabilize. The preset initial sampling time is then shifted backward along the time axis by the ringing duration to obtain the sampling point.

[0066] In one implementation, the ringing duration is the length of the ringing segment within the conduction window, and the conduction window width is the length of time the switch remains on within the current operating cycle. Both are recorded using the same timing unit, and the starting point of the window in the current operating cycle is used as a common time reference to form time data that can be proportionally converted.

[0067] Specifically, the window occupancy record consists of the duration of the ringing and the duration of the conduction window width. The window occupancy record does not change the meaning of the original time data; it only records the time domain position and length occupied by the ringing segment within the conduction window, so that the ringing duration can be read in correspondence with the conduction window width.

[0068] In one possible implementation, the controller reads the duration of the ringing from the window occupancy record, then reads the duration of the conduction window width, uses the former as the comparison time and the latter as the reference time, and obtains the time ratio through a division operation.

[0069] Specifically, let the ringing duration be T. ring The width of the conduction window is T. on The time ratio R is then calculated as R = T. ring / T on The time ratio represents the proportion of time the ringing segment occupies within the conduction window, and its value ranges from 0 to 1. The smaller the time ratio, the shorter the initial segment of the conduction window is occupied by ringing. For example, when the ringing duration is 2 microseconds and the conduction window width is 10 microseconds, the time ratio is 0.2, indicating that the ringing occupies 20% of the conduction window duration.

[0070] For example, if a ringing segment appears after the start of the conduction window, and the end time of the ringing segment is located before the conduction window, then when the time ratio is lower than a preset ratio threshold, the controller determines that the ringing noise has subsided before the conduction window. The preset ratio threshold is set as follows: Several complete electrical cycles are run under no-load or rated load conditions as calibration operating cycles. The actual waveform of the current detection path within the conduction window is recorded using an oscilloscope or sampling circuit. The ratio of the time required for the ringing to decay to the steady-state error band to the total conduction window time is calculated. The maximum value of this ratio within multiple cycles is taken as the acceptable ringing occupancy range. A margin of 10% to 20% is added to this value to obtain the preset ratio threshold, which is then written into the controller's threshold storage area. In a typical application, if the calibration measurement shows a ringing occupancy ratio of 25%, after adding a 15% margin, the preset ratio threshold is set to 40%. Further, the remaining segment of the window is the time segment within the conduction window that is after the end time of the ringing segment and does not exceed the end point of the window. When the time ratio is lower than the preset ratio threshold, the remaining segment of the window is determined to be a segment where the current tends to be stable. This determination only corresponds to the time domain state of the conduction window within the current operating cycle and does not introduce data from other operating cycles.

[0071] In one embodiment, the preset initial sampling time is a pre-configured sampling trigger reference time within the current operating cycle, located at a quarter-window position after the start of the conduction window. After obtaining the ringing duration, the controller calculates the target sampling point, which is equal to the preset initial sampling time plus the ringing duration. The controller determines whether the target sampling point is before the end time of the conduction window. If the target sampling point is within the end time of the conduction window, it is used as the final sampling point; if the target sampling point exceeds the end time of the conduction window, the time position shifted forward by 5 microseconds from the end time of the conduction window is used as the final sampling point to ensure that sampling is completed within the conduction window. The duration of the conduction window is denoted as T. on The duration of the ringing is denoted as T. ring The preset time interval between the initial sampling time and the start of the conduction window is 0.25 times T. on The time interval between the target sampling point and the start of the conduction window is 0.25 times T. on Add T ring .

[0072] Preferably, the sampling points are still limited to the width of the conduction window and correspond to the remaining segment of the window. Based on the time ratio determination result and the ringing duration shift processing, the sampling points within the current operating cycle are converted from a fixed reference time to a time point corresponding to the ringing end position. When the preset initial sampling time is shifted backward along the time axis by the same time length as the ringing duration, if the shifted sampling point does not exceed the end of the conduction window, then the shifted point is taken as the actual sampling time; if the shifted sampling point exceeds the end of the conduction window, then the sampling time is set to a preset safe time position before the end of the conduction window. The time interval between this preset safe time position and the end of the window is 5 microseconds to ensure that the sampling operation is always completed within the effective range of the conduction window.

[0073] S104. When the time ratio is not lower than the preset ratio threshold, it is determined that the ringing noise occupies too much of the conduction window and the remaining stable section is insufficient. The preset initial sampling time is delayed backward on the time axis by the width of the conduction window to obtain the sampling point.

[0074] The time ratio and a preset ratio threshold are obtained and compared under the same time reference. If the time ratio is not lower than the preset ratio threshold, a window over-occupancy flag is obtained. Based on the window over-occupancy flag, it is determined that the ringing noise occupies too much of the conduction window, and it is determined that the remaining stable segment within the conduction window is insufficient. Based on the determination that the remaining stable segment is insufficient, the conduction window width is used as the shift time, and the preset initial sampling time is delayed backward along the time axis to obtain the sampling point.

[0075] In one implementation, the time ratio represents the proportion of ringing duration within the width of the conduction window, and the preset ratio threshold represents the upper limit of the allowed ringing duration within the conduction window. The controller reads the time ratio and the preset ratio threshold within the current operating cycle and compares their values ​​on the same time reference.

[0076] Specifically, the same time reference uses the start point of the current operating cycle's conduction window as the zero point and the end point of the conduction window as the boundary. Both the time ratio and the preset ratio threshold are dimensionless values, and no new timing units are introduced. If the time ratio is less than the preset ratio threshold, the backward path of this embodiment will not be entered; if the time ratio is not less than the preset ratio threshold, a window over-occupancy flag is formed.

[0077] For example, the window over-occupancy flag is a status identifier recorded internally by the controller, and its source is solely the comparison result of the time ratio and the preset ratio threshold. The window over-occupancy flag does not replace the time ratio, nor does it change the conduction window width, but rather indicates that the ringing occupancy time in the current operating cycle has reached or exceeded the limit corresponding to the preset ratio threshold. Further, the controller reads the conduction window width in the current operating cycle based on the window over-occupancy flag and associates the conduction window width with the already determined ringing occupancy status. Since the window over-occupancy flag indicates that the ringing occupancy time has reached the ratio limit, the remaining time in the conduction window after the ringing ends is determined to be insufficient remaining stable segment. This determination only applies to the conduction window in the current operating cycle and does not extend to adjacent operating cycles.

[0078] In one possible implementation, the determination result of insufficient remaining stable segments is written into the sampling timing record of the current running cycle. The sampling timing record includes the current running cycle identifier, the conduction window width, the window over-occupancy flag, and the preset initial sampling time. The preset initial sampling time is the original sampling trigger reference, usually located after the start of the conduction window, and is used to form the time reference point before the shift.

[0079] Specifically, after determining that the remaining stable segment is insufficient, the controller no longer uses the ringing duration as the shift duration, but instead reads the conduction window width as the shift duration. The distance by which the preset initial sampling time is delayed along the time axis is equal to the conduction window width, so that the shifted time position is associated with the overall width of the current conduction window. This process is based on the time axis of the current running cycle and does not change the conduction window width itself.

[0080] In one embodiment, if the preset initial sampling time is recorded as the sampling reference point, since the ringing occupancy ratio has reached or exceeded the preset ratio threshold and the remaining stable section is insufficient, if the "preset initial sampling time plus the conduction window width" is still used as the shift result, it will exceed the conduction window. Therefore, in this case, the controller uniformly takes the sampling point as 5 microseconds ahead of the end time of the conduction window, which is a preset safety margin time. This ensures that the sampling point falls within the relatively stable section at the end of the current window, regardless of the conduction window width. The 5 microsecond safety margin is determined based on the sum of the current detection path establishment time and the sampling holding time. The controller stores the sampling point, the window over-occupancy mark, the conduction window width, and the determination result of insufficient remaining stable section in the sampling timing record of the current operating cycle. The sampling timing record is used to characterize the sampling trigger position formed when the ringing occupancy of the conduction window is too large, ensuring that effective sampling of current or voltage is completed within the current cycle.

[0081] S105. Acquire feedback signals at the sampling points, filter the feedback signals to extract the first stable current, and evaluate whether the first stable current is consistent with the second stable current obtained through the same process in the previous operating cycle.

[0082] The sampling point is acquired, and a current detection path is triggered at the sampling point to sample and acquire the feedback voltage waveform corresponding to the sampling point. The feedback voltage waveform is used as a feedback signal. The feedback signal is low-pass filtered to remove high-frequency fluctuation components, resulting in a filtered feedback signal. The first stable current of the current operating cycle is extracted from the filtered feedback signal. The second stable current obtained in the previous operating cycle using the same sampling and filtering process is acquired. The amplitude difference between the first stable current and the second stable current is compared. If the amplitude difference is lower than a preset consistency threshold, it is determined that the two are approaching consistency.

[0083] In one implementation, the sampling point is a predetermined sampling trigger time position within the current operating cycle. When the sampling point is reached, the motor drive controller triggers the current detection path to sample. The current detection path includes a phase current detection resistor, a detection amplification unit, and an analog-to-digital converter (ADC). The sampled object is the feedback voltage waveform corresponding to the sampling point. The feedback voltage waveform originates from the voltage generated by the motor winding phase current flowing through the current detection resistor. After amplitude matching, the detection amplification unit inputs the voltage to the ADC. The ADC acquires discrete voltage data within a limited sampling time near the sampling point. The discrete voltage data is arranged in the order of sampling time to form a feedback signal. The feedback voltage waveform is a continuous analog voltage signal, which is converted into a discrete digital form feedback signal after sampling and quantization by the ADC.

[0084] For example, the feedback signal is not directly used as a steady current. Since high-frequency fluctuations may still remain within the conduction window of the high-frequency switching circuit, the feedback signal includes a low-frequency variation component corresponding to the motor phase current and a high-frequency fluctuation component remaining from the switching action. Low-pass filtering is used to perform frequency separation on the feedback signal, attenuating the fluctuation components above the preset cutoff frequency and retaining the low-frequency components corresponding to the phase current changes.

[0085] In one possible implementation, the low-pass filtering process is performed using a digital low-pass filter. The digital low-pass filter receives discrete voltage data from the feedback signal, performs weighted smoothing on adjacent sampled data in chronological order, and outputs a filtered feedback signal. The weighted smoothing employs a first-order low-pass filtering algorithm, where the filtered output value of the current period equals the filtered output value of the previous period multiplied by the smoothing coefficient α, plus the sampled value of the current period multiplied by 1 minus α, where the smoothing coefficient α is calculated using the formula α = e -2π×fc×Tsfc is the preset cutoff frequency, Ts is the sampling period, and e is the base of the natural logarithm. The preset cutoff frequency fc is set according to the upper limit of the phase current fundamental frequency, typically 5 to 10 times the phase current fundamental frequency. For example, when the phase current fundamental frequency is 1kHz, fc is 5 to 10kHz.

[0086] Once set, the smoothing coefficient α remains consistent under the same motor drive conditions, ensuring that the filtering treatment is the same for the current and previous operating cycles. Furthermore, the filtered feedback signal is still in voltage form. The controller converts the filtered feedback signal into current data based on the resistance of the current sensing resistor and the amplification factor of the detection amplification unit. This conversion process can be expressed as I = U ÷ (R... s ×G), where I is the current data in amperes, U is the feedback signal voltage value obtained after detection, amplification, and filtering in volts, and R s The resistance value of the current sensing resistor is in ohms, and G is the dimensionless amplification factor. Since U is the voltage after amplification by G, it also needs to be divided by R. s The phase current can be restored by G. The resistance value R of the current detection resistor is... s The product of the current detection amplification factor G and the current detection amplification factor G is used as the detection conversion factor and written into the controller storage area after the current detection path is calibrated.

[0087] In one embodiment, the first stable current of the current operating cycle is extracted from the current data. Specifically, within a limited sampling time corresponding to the sampling point, the controller selects a stable reading segment of the filtered current data and performs averaging on the current data within this stable reading segment to obtain the first stable current. The method for determining the stable reading segment is as follows: the controller calculates the absolute value of the current difference between two adjacent sampling points, |I(n)-I(n-1)|, where I(n) is the current value of the nth sampling point and I(n-1) is the current value of the (n-1)th sampling point. When the absolute value of the current difference between 20 or more consecutive sampling points is less than the stability determination threshold, the data segment consisting of these consecutive sampling points is identified as a stable reading segment. The stability determination threshold is set based on the allowable range of quantization noise and phase current ripple of the analog-to-digital converter unit, typically 0.05 amperes. If multiple stable reading segments exist within the limited sampling time, the stable reading segment with the longest duration is selected for averaging. The controller establishes a sampling record for each operating cycle. This record includes the operating cycle identifier, sampling time, current detection path number, low-pass filter parameters, and extracted steady-state current value. The operating cycle identifier uses an incrementing integer numbering method, with an initial value of 1, and automatically increments by 1 after each operating cycle. The sampling records are stored in the controller's non-volatile memory, and an index table is established according to the operating cycle identifier. The controller can read the sampling record of the previous operating cycle by decrementing 1 from the current operating cycle identifier.

[0088] Preferably, the second stable current of the previous operating cycle is obtained according to the same sampling point determination process, the same current detection path sampling process, the same low-pass filtering process, and the same stable current extraction process. After reading the sampling record of the previous operating cycle, the controller extracts the stored second stable current value and compares it with the first stable current of the current operating cycle.

[0089] Specifically, after reading the first stable current and the second stable current, the controller compares their amplitude difference under the same current unit. The amplitude difference is the absolute value of the difference between the first stable current and the second stable current, i.e. ,in The first stable current, This is the second steady current, and the amplitude difference is used to characterize the degree of deviation between steady current readings in adjacent operating cycles.

[0090] In one implementation, a preset consistency threshold is used to define the allowable deviation range of the steady-state current readings between adjacent operating cycles. The preset consistency threshold is determined based on the sampling fluctuation statistics of the motor phase current detection path under stable operating conditions. Specifically, it involves continuously collecting steady-state current data for several cycles under steady-state motor operation and calculating the standard deviation of the absolute value of the difference between the steady-state currents of adjacent cycles. Take a preset consistency threshold equal to 3 times. The typical value is 0.05 amperes, and it is written into the controller's threshold storage area. When the amplitude difference is lower than a preset consistency threshold, the controller determines that the first stable current and the second stable current tend to be consistent.

[0091] It should be noted that if the current operating cycle is the first sampleable operating cycle after startup, and the second stable current of the previous operating cycle has not yet been formed, the controller only establishes the current sampling record as the reference data after the first stable current is formed. When the second operating cycle arrives, the controller performs a consistency judgment based on the stored sampling record of the first cycle and the first stable current of the current cycle to determine whether the sampling point falls within the current stable region. If the judgment passes, the controller outputs the target feedback signal corresponding to the second cycle. If the judgment fails, a new sampling record is established and the judgment process is repeated in the next cycle. This process is still limited to the phase current feedback sampling process of motor drive under high-frequency switching circuit.

[0092] S106. When the current tends to be consistent, it is determined that the sampling point falls into the current stable region. The first stable current is output as the target feedback signal to the closed-loop control module to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

[0093] The consistency judgment result of the first stable current and the second stable current is obtained. If the judgment result is consistent, a current stability zone marker is obtained based on the sampling point. The first stable current is read according to the current stability zone marker and used as the target feedback signal to obtain target feedback data with the current operating cycle identifier. The target feedback signal is output to the closed-loop control module through the target feedback data. The closed-loop control module calculates the pulse width modulation correction amount based on the deviation between the target feedback signal and the target torque command, and adjusts the switching timing of the high-frequency switching circuit based on the pulse width modulation correction amount to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

[0094] In one implementation, the convergence judgment result is derived from a comparison of the amplitude difference between the first stable current of the current operating cycle and the second stable current of the previous operating cycle. When the convergence judgment result indicates that the two are within a preset convergence threshold range, the controller records the sampling point as a stable sampling position within the current operating cycle and forms a current stability zone marker.

[0095] Specifically, the current stabilization zone marker is used to indicate that the current detection path output corresponding to the sampling point has entered a stable reading range. The current stabilization zone marker is recorded together with the current operating cycle identifier, the sampling point, and the first stable current, so that the source of the target feedback signal is consistent with the corresponding operating cycle, and the reading within the ringing noise range is not used as the feedback source.

[0096] In one possible implementation, the controller reads the first stable current based on the current stability zone marker. The process of obtaining the first stable current is as follows: first, the feedback signal is filtered to obtain a filtered feedback signal; then, the filtered feedback signal is converted into a current value using a current conversion formula; finally, a stable reading segment is extracted from the converted current value, and the unit of this current value is consistent with the calibration unit of the motor phase current detection path. The controller no longer reads the original feedback voltage waveform, but instead uses the first stable current as the target feedback signal.

[0097] For example, the target feedback signal and the target feedback data are not the same object. The target feedback signal is the current feedback value participating in the closed-loop control, and the target feedback data is the recorded data carrying the target feedback signal. The target feedback data includes at least the current operating cycle identifier, sampling point, current stability zone marker, and target feedback signal, enabling the closed-loop control module to read the corresponding feedback values ​​in cycle order. During the system initialization phase, the motor drive controller reads the motor parameter table from the memory. This parameter table is obtained through bench calibration before leaving the factory and records the actual output torque values ​​corresponding to different current values, forming a discrete current-torque mapping relationship. During the closed-loop control process, the conversion calculation is completed by looking up the table or linear interpolation. Further, after receiving the target feedback data, the closed-loop control module extracts the target feedback signal from it and reads the target torque command. The target torque command is the torque setpoint given by the motor drive controller in the current operating cycle, and the target feedback signal is used to reflect the actual current state of the motor windings in the current operating cycle. The closed-loop control module converts the target feedback signal into a feedback torque amount according to the current-torque mapping relationship in the motor parameter table.

[0098] In one embodiment, the closed-loop control module compares the feedback torque with the target torque command to obtain torque deviation information. The torque deviation information indicates the direction and magnitude of the difference between the target torque command and the feedback torque. If the feedback torque is lower than the target torque command, the pulse width modulation (PWM) correction is increased accordingly; if the feedback torque is higher than the target torque command, the PWM correction is decreased accordingly. The PWM correction is a PWM duty cycle adjustment calculated by the closed-loop control module based on the torque deviation. The drive controller adds this correction to the current PWM duty cycle and updates the conduction time of the next control cycle, ensuring that the conduction cycle of the high-frequency switching circuit's bridge arm matches the current demand corresponding to the target torque command. For example, if the current PWM duty cycle is 60% and the correction is +5%, the PWM duty cycle of the next control cycle is updated to 65%.

[0099] Preferably, the target feedback data undergoes a periodic consistency check before being output to the closed-loop control module. The check includes: whether the current operating cycle identifier matches the cycle identifier of the closed-loop control module, and whether the sampling point carries the current stable region marker. After the check passes, the target feedback signal enters the closed-loop control module to form a pulse width modulation correction amount.

[0100] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. A method for coordinated control of motor drive under a high-frequency switching circuit, characterized in that, include: Obtain the duty cycle of the PWM signal used by the high-frequency switching circuit to drive the motor, and determine the conduction window width of the high-frequency switching circuit in the current operating cycle based on the signal duty cycle; Collect ringing noise signals within the width of the conduction window, evaluate whether the peak voltage of the ringing noise signals is higher than a preset voltage threshold, and determine that there is effective ringing noise within the conduction window when the peak voltage is higher than the preset voltage threshold. Then, extract the duration of the ringing noise signals to obtain the ringing duration. Calculate the proportion of the ringing duration within the width of the conduction window. When the proportion is lower than a preset proportion threshold, it is determined that the ringing noise has subsided in the first part of the conduction window and the current in the remaining section of the window tends to be stable. The preset initial sampling time is delayed backward on the time axis by the ringing duration to obtain the sampling point. When the time ratio is not lower than the preset ratio threshold, it is determined that the ringing noise occupies too much of the conduction window and there is not enough remaining stable section. The preset initial sampling time is delayed backward on the time axis by the width of the conduction window to obtain the sampling point. A feedback signal is collected at the sampling point, the feedback signal is filtered to extract the first stable current, and the consistency between the first stable current and the second stable current obtained by the same process in the previous running cycle is evaluated. When the sampling points converge, it is determined that they fall into the current stable region. The first stable current is then output as the target feedback signal to the closed-loop control module to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

2. The motor drive coordinated control method under a high-frequency switching circuit according to claim 1, characterized in that, The step of obtaining the duty cycle of the PWM signal used by the high-frequency switching circuit to drive the motor, and determining the conduction window width of the high-frequency switching circuit in the current operating cycle based on the signal duty cycle, includes: Obtain PWM signal duty cycle information, and determine the conduction window width within the current operating cycle based on the correspondence between the duty cycle information and the switching cycle; The conduction window segment is divided based on the conduction window width, and a conduction window segment description is formed by combining the duty information and the relationship between the switching cycle. Consistency processing is performed on the description of the conduction window segment, and the conduction window width result is obtained based on the conduction window width.

3. The motor drive coordinated control method under a high-frequency switching circuit according to claim 1, characterized in that, The ringing noise signal within the width of the acquisition conduction window is collected. The peak voltage of the ringing noise signal is evaluated to determine if it exceeds a preset voltage threshold. If the peak voltage exceeds the preset voltage threshold, it is determined that valid ringing noise exists within the conduction window. The duration of the ringing noise signal is then extracted to obtain the ringing duration, including: Obtain the window start point and window end point corresponding to the width of the conduction window, collect the transient voltage waveform of the current detection path between the window start point and the window end point, and extract the ringing noise signal from the transient voltage waveform; For the ringing noise signal, the voltage amplitude of each sampling point is read, and the largest amplitude value is selected from the voltage amplitude as the peak voltage. When the peak voltage is higher than the preset voltage threshold, an over-limit ringing mark is obtained. The continuous over-limit segment formed by locating adjacent over-limit sampling points based on the over-limit ringing mark is determined by the starting and ending boundaries of the continuous over-limit segment, and the ringing duration is obtained.

4. The motor drive coordinated control method under a high-frequency switching circuit according to claim 1, characterized in that, The calculation of the ringing duration as a proportion of the time within the conduction window width is used. When the proportion is lower than a preset threshold, it is determined that the ringing noise has subsided in the first part of the conduction window and the current in the remaining section of the window tends to stabilize. The sampling point is obtained by delaying the preset initial sampling time backward on the time axis using the ringing duration, including: Obtain the ringing duration and the conduction window width, and extract the duration of the ringing duration and the duration of the conduction window width according to the same time base to form a window occupancy record; Based on the window occupancy record, the duration of the ringing is proportionally converted to the duration of the conduction window width to obtain the time ratio; If the time ratio is lower than the preset ratio threshold, it is determined that the ringing noise subsides in the first part of the conduction window and the current in the remaining section of the window tends to be stable. The preset initial sampling time is then shifted back along the time axis by the duration of the ringing noise to obtain the sampling point.

5. The motor drive coordinated control method under a high-frequency switching circuit according to claim 1, characterized in that, When the time ratio is not lower than a preset ratio threshold, it is determined that the ringing noise occupies too large a conduction window and the remaining stable segment is insufficient. The sampling point is obtained by delaying the preset initial sampling time backward on the time axis using the conduction window width, including: Obtain the time ratio and a preset ratio threshold, compare their sizes under the same time base, and if the time ratio is not lower than the preset ratio threshold, then obtain a window over-occupancy flag; Based on the window over-occupancy marker, it is determined that the ringing noise occupies too much of the conduction window, and that the remaining smooth section within the conduction window is insufficient. Based on the determination that the remaining stable section is insufficient, the width of the conduction window is used as the shift time, so that the preset initial sampling time is delayed backward along the time axis to obtain the sampling point.

6. The motor drive coordinated control method under a high-frequency switching circuit according to claim 1, characterized in that, The process of acquiring feedback signals at the sampling points, filtering the feedback signals to extract a first stable current, and evaluating whether the first stable current is consistent with the second stable current obtained through the same process in the previous operating cycle includes: The sampling point is obtained, and the current detection path is triggered at the sampling point to perform sampling. The feedback voltage waveform corresponding to the sampling point is collected, and the feedback voltage waveform is used as the feedback signal. The feedback signal is low-pass filtered to remove high-frequency fluctuation components, resulting in a filtered feedback signal. The first stable current of the current operating cycle is then extracted from the filtered feedback signal. The second stable current obtained from the previous running cycle using the same sampling and filtering process is acquired. The amplitude difference between the first stable current and the second stable current is compared. If the amplitude difference is lower than a preset consistency threshold, it is determined that the two are approaching consistency.

7. The motor drive coordinated control method under a high-frequency switching circuit according to claim 1, characterized in that, When the sampling points converge, it is determined that they fall into the current stability region. The first stable current is then output as the target feedback signal to the closed-loop control module to complete the torque coordinated control of the motor drive under the high-frequency switching circuit, including: Obtain the result of the first stable current and the second stable current converging. If the result is converging, then obtain the current stable region marker based on the sampling point. The first stable current is read according to the current stable region mark, and the first stable current is used as the target feedback signal to obtain target feedback data with the current operating cycle identifier. The target feedback data is used to output the target feedback signal to the closed-loop control module. The closed-loop control module calculates the pulse width modulation correction amount based on the deviation between the target feedback signal and the target torque command, and adjusts the switching timing of the high-frequency switching circuit based on the pulse width modulation correction amount to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

8. The motor drive coordinated control method under a high-frequency switching circuit according to claim 3, characterized in that, The voltage amplitude of each sampling point of the ringing noise signal is read, and the largest amplitude value is selected as the peak voltage. When the peak voltage is higher than a preset voltage threshold, an over-limit ringing flag is obtained, including: The continuous over-limit segment formed by locating adjacent over-limit sampling points based on the over-limit ringing mark is determined by the starting and ending boundaries of the continuous over-limit segment, and the ringing duration is obtained.

9. The motor drive coordinated control method under a high-frequency switching circuit according to claim 6, characterized in that, The step of performing low-pass filtering on the feedback signal to remove high-frequency fluctuation components in the feedback signal to obtain a filtered feedback signal, and extracting the first stable current of the current operating cycle from the filtered feedback signal, includes: The second stable current obtained from the previous running cycle using the same sampling and filtering process is acquired. The amplitude difference between the first stable current and the second stable current is compared. If the amplitude difference is lower than a preset consistency threshold, it is determined that the two are approaching consistency.

10. A motor drive coordinated control method under a high-frequency switching circuit according to claim 7, characterized in that, The step of reading the first stable current according to the current stable region marker and using the first stable current as the target feedback signal to obtain target feedback data with the current operating cycle identifier includes: The target feedback data is used to output the target feedback signal to the closed-loop control module. The closed-loop control module calculates the pulse width modulation correction amount based on the deviation between the target feedback signal and the target torque command, and adjusts the switching timing of the high-frequency switching circuit based on the pulse width modulation correction amount to complete the torque coordinated control of the motor drive under the high-frequency switching circuit.

Images