Micro-motor stall protection method and system based on current waveform analysis
By performing variational mode decomposition and Hilbert transform on the micro-motor current signal, an electromechanical deviation index is constructed, which solves the misjudgment problem of existing micro-motor stall protection methods and realizes fast and reliable stall protection and equipment safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN HIGH PRECISION MOTOR CO LTD
- Filing Date
- 2026-03-19
- Publication Date
- 2026-06-12
Smart Images
Figure CN122203151A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of micro motors, and in particular relates to a method and system for locking rotor protection of micro motors based on current waveform analysis. Background Technology
[0002] Micro motors are widely used in industrial automation, smart homes, medical devices, and consumer electronics. However, in actual operation, micro motors may stall due to mechanical jamming, excessive load, or transmission mechanism failure. Stalling is a serious fault condition where the motor rotor cannot rotate, but the stator windings are still energized, preventing the input electrical energy from being converted into mechanical energy. Almost all of this energy is dissipated as heat in the windings. This causes a sharp increase in winding current and a rapid rise in temperature. If the power is not cut off in time, the motor windings can easily burn out in a short period, damaging the drive circuit and even causing fires or other safety accidents. Existing micro motor protection technologies are mainly divided into several categories. The most traditional method is threshold protection based on current or temperature. Overcurrent protection determines the fault by detecting whether the motor current exceeds a fixed threshold. However, this method is difficult to distinguish between stall faults and high currents under normal operating conditions such as motor starting or sudden load increases. Setting the threshold too low can easily lead to false alarms, while setting it too high will result in untimely protection. Temperature protection is achieved by installing temperature sensors inside or on the surface of the motor. However, due to thermal inertia, the response speed is slow, and the motor often suffers irreversible damage by the time overheating is detected. When zero speed is detected and the current rises abnormally, it is judged as stalled. Although this method is direct and effective, it adds extra hardware sensors, increases system cost and complexity, reduces reliability, and is not suitable for sensorless control applications. In addition, there are some sensorless detection methods based on signal processing, which identify stalled by analyzing the spectrum or statistical characteristics of the current signal. However, these methods are prone to misjudging heavy load conditions as stalled, and their robustness and diagnostic accuracy still need to be improved. In particular, they have not effectively established a comprehensive diagnostic index that can simultaneously reflect the mismatch between dynamic power input and mechanical motion state. Summary of the Invention
[0003] To address the aforementioned problems, in the first aspect, this invention proposes a micromotor stall protection method based on current waveform, comprising the following steps:
[0004] S1, acquire the current signal of the stator winding during micro-motor operation as raw data;
[0005] S2, perform variational mode decomposition on the original data to decompose it into intrinsic mode functions (IMFs) of a preset target number of K modes; based on the center frequency of each IMF component, select the fundamental current IMF component representing the power grid fundamental frequency, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics.
[0006] S3. Perform a Hilbert transform on the fundamental current IMF component, extract its amplitude envelope, and differentiate it with respect to time to obtain the rate of change of electrical energy injection representing the instantaneous input power change; perform a Hilbert transform on the specific high-frequency IMF component, extract its instantaneous frequency, differentiate it with respect to time, and then invert it to obtain the virtual mechanical deceleration representing the change of motor speed; combine the rate of change of electrical energy injection and the virtual mechanical deceleration to obtain the electromechanical deviation degree;
[0007] S4. Compare the currently calculated electromechanical deviation with the dynamic preset threshold. If the electromechanical deviation continues to exceed the dynamic preset threshold, it is determined that the micro motor is stalled.
[0008] S5, upon determining that a stall has occurred, immediately blocks the PWM drive signal of the drive bridge and controls the drive bridge to enter the active freewheeling mode to quickly dissipate the inductance energy stored in the stator winding, thereby protecting the micro motor.
[0009] Optionally, the preset target mode number K is an optimal integer value that can effectively separate the fundamental current from specific high-frequency harmonics, determined in advance based on the electrical characteristics and mechanical structure of the micro-motor under test through offline experiments or simulation analysis.
[0010] Optionally, the step of filtering out the fundamental current IMF component representing the grid fundamental frequency and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics based on the center frequency of each IMF component includes:
[0011] Calculate the center frequencies of K intrinsic mode functions (IMFs); take the IMF component whose center frequency is closest to the power supply frequency of the micromotor as the fundamental current IMF component; take one or more IMF components with high energy in a specific harmonic frequency band as specific high-frequency IMF components.
[0012] Optionally, the step of performing a Hilbert transform on the fundamental current IMF component, extracting its amplitude envelope, and differentiating it with respect to time to obtain the rate of change of power injection representing the instantaneous change in input power includes:
[0013] The fundamental current IMF component is subjected to Hilbert transform to obtain an analytical signal; the magnitude of the analytical signal is calculated to obtain the transient amplitude envelope; the rate of change of electrical energy injection is obtained by calculating the difference between the transient amplitude envelope at the current sampling time and the previous sampling time.
[0014] Optionally, the step of performing a Hilbert transform on the specific high-frequency IMF component, extracting its instantaneous frequency, and then inverting the derivative with respect to time to obtain the virtual mechanical deceleration representing the change in motor speed includes:
[0015] If multiple specific high-frequency IMF components are selected, they are superimposed into a composite high-frequency signal; Hilbert transform is performed on the specific high-frequency IMF components or the composite high-frequency signal to obtain an analytical signal; the phase of the analytical signal is calculated and unwrapped; the unwrapped phase is subjected to first-order difference to obtain the instantaneous frequency; the instantaneous frequency is then subjected to first-order difference, and the result is inverted to obtain the virtual mechanical deceleration.
[0016] Optionally, the step of combining the rate of change of electrical energy injection with the virtual mechanical deceleration to obtain the electromechanical deviation includes:
[0017] The electromechanical deviation is obtained by multiplying the rate of change of electrical energy injection by the virtual mechanical deceleration.
[0018] Optionally, the method further includes:
[0019] Under stable operation of the micro motor, electromechanical deviation data for N sampling periods are continuously collected to form a sliding window; the moving average and moving standard deviation of the data within the sliding window are calculated in real time; the moving average and the moving standard deviation are added together and the sum is used as the dynamic preset threshold.
[0020] In a second aspect, the present invention proposes a micro-motor stall protection system based on current waveform, comprising the following modules:
[0021] The acquisition module is used to acquire the current signal of the stator winding of the micro motor as raw data during operation;
[0022] The decomposition module is used to perform variational mode decomposition on the original data, decomposing it into intrinsic mode functions (IMFs) of a preset target number of K modes; based on the center frequency of each IMF component, the fundamental current IMF component representing the power grid fundamental frequency, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics are selected.
[0023] The calculation module is used to perform a Hilbert transform on the fundamental current IMF component, extract its amplitude envelope, and differentiate it with respect to time to obtain the rate of change of power injection representing the instantaneous input power change; to perform a Hilbert transform on the specific high-frequency IMF component, extract its instantaneous frequency, differentiate it with respect to time, and then invert it to obtain the virtual mechanical deceleration representing the change of motor speed; and to combine the rate of change of power injection and the virtual mechanical deceleration to obtain the electromechanical deviation degree.
[0024] The comparison module is used to compare the currently calculated electromechanical deviation with a dynamic preset threshold. If the electromechanical deviation continues to exceed the dynamic preset threshold, it is determined that the micro motor is stalled.
[0025] The protection module is used to immediately block the PWM drive signal of the drive bridge and control the drive bridge to enter the active freewheeling mode after a stall occurs, so as to quickly dissipate the inductance energy stored in the stator winding and protect the micro motor.
[0026] Optionally, the preset target mode number K is an optimal integer value that can effectively separate the fundamental current from specific high-frequency harmonics, determined in advance based on the electrical characteristics and mechanical structure of the micro-motor under test through offline experiments or simulation analysis.
[0027] Optionally, the step of filtering out the fundamental current IMF component representing the grid fundamental frequency and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics based on the center frequency of each IMF component includes:
[0028] Calculate the center frequencies of K intrinsic mode functions (IMFs); take the IMF component whose center frequency is closest to the power supply frequency of the micromotor as the fundamental current IMF component; take one or more IMF components with high energy in a specific harmonic frequency band as specific high-frequency IMF components.
[0029] Optionally, the step of performing a Hilbert transform on the fundamental current IMF component, extracting its amplitude envelope, and differentiating it with respect to time to obtain the rate of change of power injection representing the instantaneous change in input power includes:
[0030] The fundamental current IMF component is subjected to Hilbert transform to obtain an analytical signal; the magnitude of the analytical signal is calculated to obtain the transient amplitude envelope; the rate of change of electrical energy injection is obtained by calculating the difference between the transient amplitude envelope at the current sampling time and the previous sampling time.
[0031] Optionally, the step of performing a Hilbert transform on the specific high-frequency IMF component, extracting its instantaneous frequency, and then inverting the derivative with respect to time to obtain the virtual mechanical deceleration representing the change in motor speed includes:
[0032] If multiple specific high-frequency IMF components are selected, they are superimposed into a composite high-frequency signal; Hilbert transform is performed on the specific high-frequency IMF components or the composite high-frequency signal to obtain an analytical signal; the phase of the analytical signal is calculated and unwrapped; the unwrapped phase is subjected to first-order difference to obtain the instantaneous frequency; the instantaneous frequency is then subjected to first-order difference, and the result is inverted to obtain the virtual mechanical deceleration.
[0033] Optionally, the step of combining the rate of change of electrical energy injection with the virtual mechanical deceleration to obtain the electromechanical deviation includes:
[0034] The electromechanical deviation is obtained by multiplying the rate of change of electrical energy injection by the virtual mechanical deceleration.
[0035] Optionally, the method further includes:
[0036] Under stable operation of the micro motor, electromechanical deviation data for N sampling periods are continuously collected to form a sliding window; the moving average and moving standard deviation of the data within the sliding window are calculated in real time; the moving average and the moving standard deviation are added together and the sum is used as the dynamic preset threshold.
[0037] This invention utilizes variational mode decomposition to separate the fundamental component reflecting the grid's base frequency from the specific high-frequency component of the motor's rotational speed, thereby improving the signal processing's anti-interference capability and the accuracy of feature extraction. An electromechanical deviation index is constructed, simultaneously reflecting the rate of change in instantaneous input power and the virtual mechanical deceleration reflecting changes in motor speed. This index measures the imbalance between electrical and mechanical quantities during motor stall, making stall fault identification more sensitive and reliable. It avoids the misjudgment or missed judgment problems caused by traditional methods relying solely on a single current amplitude, which are susceptible to load fluctuations and transient disturbances. Once stall is detected, the drive can be quickly cut off and the system can enter active freewheeling mode, achieving rapid response and effective protection for the motor and drive circuit, thus extending the equipment's service life. Attached Figure Description
[0038] Figure 1 A flowchart of the first embodiment;
[0039] Figure 2 A schematic diagram of the rate of change of electrical energy injection;
[0040] Figure 3 This is a schematic diagram of the degree of divergence. Detailed Implementation
[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0042] Figure 1 The flowchart for the first embodiment of the present invention is shown below. Figure 1 This includes the following steps:
[0043] S1, acquire the current signal of the stator winding during the operation of the micro motor as the raw data.
[0044] The analog current signal of the stator winding of the micro motor is acquired by a Hall current sensor or a sampling resistor connected in series on the bus or bridge arm of the drive circuit; the analog signal is sampled by a high-speed analog-to-digital converter (ADC) with the sampling frequency set to an integer multiple of the switching frequency, such as 10kHz, to convert the continuous analog signal into a discrete digital time series as the raw data stream.
[0045] S2, perform variational mode decomposition on the original data to decompose it into intrinsic mode functions (IMFs) of a preset target number of K modes; based on the center frequency of each IMF component, select the fundamental current IMF component representing the power grid fundamental frequency, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics.
[0046] Call the Variational Mode Decomposition (VMD) algorithm and set the penalty factor. Hyperparameters such as the number of modes K are used. For example, the value of K is preset to 5 based on the complexity of the motor spectrum. The VMD algorithm is executed on the current time series obtained by S1 to obtain K IMF components and their corresponding center frequencies. All IMF components are traversed, and the IMF component with the center frequency closest to the power grid frequency of 50Hz is determined as the fundamental current IMF component. The IMF components with center frequencies falling within the characteristic frequency range calculated from the motor slot frequency or rotational frequency harmonics are determined as specific high-frequency IMF components.
[0047] S3. Perform a Hilbert transform on the fundamental current IMF component, extract its amplitude envelope, and differentiate it with respect to time to obtain the rate of change of electrical energy injection representing the instantaneous change of input power; perform a Hilbert transform on the specific high-frequency IMF component, extract its instantaneous frequency, differentiate it with respect to time, and invert it to obtain the virtual mechanical deceleration representing the change of motor speed; combine the rate of change of electrical energy injection with the virtual mechanical deceleration to obtain the electromechanical deviation.
[0048] The fundamental current IMF component is processed using a Hilbert transform algorithm based on Fast Fourier Transform (FFT), such as the `scipy.signal.hilbert` function, to obtain an analytic signal. The magnitude of this analytic signal is then calculated using the `numpy.abs` function to obtain the amplitude envelope of the fundamental current. A first-order backward difference method is then used to numerically differentiate the amplitude envelope sequence to obtain the rate of change of electrical energy injection, such as... Figure 2As shown. The Hilbert transform is also applied to specific high-frequency IMF components to obtain analytic signals; the phase angle of the analytic signal is calculated using the arctangent function atan2, and phase jumps are processed using a phase expansion algorithm such as numpy.unwrap to obtain the instantaneous phase. The instantaneous frequency is obtained by numerically differentiating the instantaneous phase sequence using the first-order backward difference method; the virtual mechanical deceleration is obtained by differentiating the instantaneous frequency sequence again using the first-order backward difference method and taking its inverse; the rate of change of electrical energy injection is multiplied point-by-point with the virtual mechanical deceleration to obtain the electromechanical deviation time series, as shown. Figure 3 As shown.
[0049] S4. Compare the currently calculated electromechanical deviation with the dynamic preset threshold. If the electromechanical deviation continues to exceed the dynamic preset threshold, it is determined that the micro motor has stalled.
[0050] A sliding window mean filtering algorithm is used to calculate the average value of the electromechanical deviation degree sequence under normal operating conditions over a period of time. This average value is multiplied by a coefficient greater than 1, and a fixed safety margin is added to obtain the current dynamic preset threshold. In each sampling period, the latest electromechanical deviation degree value calculated by S3 is compared with this dynamic threshold. A fault duration counter is set; if the current electromechanical deviation degree value is greater than the dynamic threshold, the counter is incremented; if it is less than or equal to the dynamic threshold, the counter is reset to zero. When the cumulative counter value reaches a preset judgment time length, for example, 50 milliseconds, a stall fault is determined to have occurred. Optionally, a start-up shield timer is provided to temporarily suppress the output of the stall judgment result within a predetermined time, such as 200ms, after the micromotor receives the start command, to prevent false protection caused by the huge starting current and low speed at the moment of motor start-up.
[0051] S5, upon determining that a stall has occurred, immediately blocks the PWM drive signal of the drive bridge and controls the drive bridge to enter the active freewheeling mode to quickly dissipate the inductance energy stored in the stator winding, thereby protecting the micro motor.
[0052] The stall fault detection signal triggers the microcontroller's (MCU) interrupt service routine. In this routine, instructions are first written to the relevant registers of the PWM controller. For example, for the STM32 series microcontroller, by setting the MOE bit in the BDTR register of the TIM advanced timer to 0, the PWM output of all upper and lower bridge arm power transistors of the drive bridge is immediately turned off, thus blocking the drive signal. Subsequently, by operating the GPIO port register, the gate drive signals of all lower bridge arm power transistors of the drive bridge are forcibly set to high level, while the gate drive signals of all upper bridge arm power transistors are set to low level, so that the drive bridge enters the active freewheeling state with all lower transistors conducting, providing a low-impedance attenuation loop for the stator winding current.
[0053] In a preferred embodiment, the preset target mode number K is an optimal integer value that can effectively separate the fundamental current from specific high-frequency harmonics, determined in advance based on the electrical characteristics and mechanical structure of the micro-motor under test through offline experiments or simulation analysis.
[0054] To determine the optimal number of modes K, comprehensive offline data acquisition and analysis of the target micromotor are required. On a motor testing platform, the stator current waveforms are recorded under various steady-state and transient conditions, including start-up, no-load, rated load, overload, and artificially induced slow stall and emergency stop stall. The sampling frequency must satisfy the Nyquist theorem, preferably 10kHz to 50kHz. Iterative variational mode decomposition (VMD) tests are performed on the acquired data. A search range for the K value is set; for example, for common micro permanent magnet synchronous motors, the K value can range from 3 to 10. For each K value, the VMD algorithm is executed, and the decomposition effect is evaluated. The evaluation criteria are mainly twofold: first, the separation degree of the fundamental current IMF component, i.e., whether there exists an IMF component whose center frequency is tightly locked to the grid fundamental frequency, for example, around 50Hz ± 1Hz, and whose shape is smooth with no obvious mode aliasing. Secondly, the clarity of key harmonic IMF components is crucial. This involves estimating the frequency range of characteristic harmonics related to rotational speed or cogging frequency using the number of pole pairs, slots, and rotational speed of the motor. For example, can 150Hz to 800Hz be stably decomposed into one or two independent IMF components, rather than being scattered across multiple components or mixed with noise? For instance, testing a 4-pole, 12-slot permanent magnet synchronous micromotor revealed that at K=5, the fundamental frequency and rotational harmonics were not completely separated; while at K=7, over-decomposition occurred, splitting a single harmonic component into two IMFs. Ultimately, it was found that when K was set to 6, the 50Hz fundamental frequency, the main rotational harmonics, and other noise components could be most clearly distinguished. Therefore, the preset target mode number K for this motor was set to 6. Simultaneously, to ensure the convergence and accuracy of the decomposition, a quadratic penalty factor in the VMD algorithm was used. The fidelity parameter is typically set between 1000 and 3000, such as 2000. Set it to 0.
[0055] In a preferred embodiment, the step of filtering out the fundamental current IMF component representing the grid fundamental frequency and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics, based on the center frequency of each IMF component, includes:
[0056] Calculate the center frequencies of K intrinsic mode functions (IMFs); take the IMF component whose center frequency is closest to the power supply frequency of the micromotor as the fundamental current IMF component; take one or more IMF components with high energy in a specific harmonic frequency band as specific high-frequency IMF components.
[0057] After VMD decomposition, the system obtains K IMF components and their corresponding center frequencies. First, automatic filtering of the fundamental component is performed: the preset power grid reference frequency is selected. For example, China's center frequency is 50.0 Hz, which is the center frequency of all IMF components. Compare values from (k=1,...,K) and select the absolute value of the difference. The smallest IMF component is used as the fundamental current. For example, if the six decomposed center frequencies are 15.2Hz, 49.9Hz, 245.6Hz, 480.1Hz, 950.8Hz, and 1800.3Hz, then the IMF component with a center frequency of 49.9Hz is identified as the fundamental frequency. Specific high-frequency IMF components are then screened. Based on the motor's design parameters, such as the number of pole pairs (p), the number of slots (Z), and the rated speed range, one or more harmonic frequency bands are pre-defined. For example, a speed-related harmonic frequency band can be set to [150Hz, 600Hz]. All non-fundamental IMF components are iterated through to find all candidate components whose center frequencies fall within this frequency band. In this example, the two IMF components with center frequencies of 245.6Hz and 480.1Hz become candidates. To enhance the robustness of the signal, the sum of the squares of the energy amplitudes of these two candidate components within the most recent short time window, such as 10 milliseconds, can be further calculated, and the component with the highest energy proportion is selected as the dominant specific high-frequency IMF component. In some implementations, if the energies of multiple candidate components are all high and strongly correlated with the rotational speed change, they can be linearly superimposed to obtain a composite high-frequency signal. ,in , These are the weighting coefficients.
[0058] In a preferred embodiment, performing a Hilbert transform on the fundamental current IMF component, extracting its amplitude envelope, and differentiating it with respect to time to obtain the rate of change of power injection representing the instantaneous change in input power includes:
[0059] The fundamental current IMF component is subjected to Hilbert transform to obtain an analytical signal; the magnitude of the analytical signal is calculated to obtain the transient amplitude envelope; the rate of change of electrical energy injection is obtained by calculating the difference between the transient amplitude envelope at the current sampling time and the previous sampling time.
[0060] Obtain the fundamental current component Then, its orthogonal components are calculated using the standard Hilbert transform algorithm. Thus constructing an analytical signal .
[0061] Calculate the magnitude, i.e., the instantaneous amplitude, of the analytic signal at each sampling time n. Amplitude envelope Represents the peak value variation of the fundamental current. Rate of change of electrical energy injection. The rate of change of the instantaneous amplitude envelope with respect to time is approximated by performing a first-order backward difference. ,in The sampling period. In practical applications, due to... Since this is a constant, and the subsequent comparison is with a dynamic threshold, it can be omitted to simplify the calculation. For example, if the motor current sampling frequency is 10kHz... At the moment of stall, the fundamental current amplitude may jump from 2.0A to 2.2A within 0.1ms, then the calculated... It is a positive value of 0.2.
[0062] In a preferred embodiment, the step of performing a Hilbert transform on the specific high-frequency IMF component, extracting its instantaneous frequency, and then inverting the derivative with respect to time to obtain the virtual mechanical deceleration representing the change in motor speed includes:
[0063] If multiple specific high-frequency IMF components are selected, they are superimposed into a composite high-frequency signal; Hilbert transform is performed on the specific high-frequency IMF components or the composite high-frequency signal to obtain an analytical signal; the phase of the analytical signal is calculated and unwrapped; the unwrapped phase is subjected to first-order difference to obtain the instantaneous frequency; the instantaneous frequency is then subjected to first-order difference, and the result is inverted to obtain the virtual mechanical deceleration.
[0064] Select a specific high-frequency component Alternatively, after processing the composite signal, perform a Hilbert transform on it to obtain the analytic signal. Calculate its instantaneous phase. Because the range of values for the arctangent function is limited to... Between, the calculated phase sequence exists The transition is so rapid that it must be unwrapped to obtain a continuously monotonically changing phase sequence. Instantaneous frequency The phase after unwinding is obtained by performing a first-order difference and scaling transformation, i.e. Since frequency is directly proportional to motor speed, its rate of change reflects the motor's angular acceleration. Virtual mechanical deceleration. The following is obtained by performing a first-order backward difference on the instantaneous frequency sequence and then inverting it: When the motor suddenly stalls, the speed drops sharply, leading to... much smaller , It is a large negative value, therefore This is expressed as a significantly positive value. To simplify calculations, the constant denominator... It can be omitted because its function is equivalent to a fixed scaling factor and does not affect the comparison result with the dynamic threshold.
[0065] In a preferred embodiment, the step of combining the rate of change of electrical energy injection with the virtual mechanical deceleration to obtain the electromechanical deviation includes:
[0066] The electromechanical deviation is obtained by multiplying the rate of change of electrical energy injection by the virtual mechanical deceleration.
[0067] A preferred implementation is to multiply the two: This greatly amplifies the stalled characteristics. When stalling occurs, the controller commands the injection of more current to overcome the large load. The value is a large positive value, while at the same time the motor rotor is decelerating rapidly, leading to... Also a large positive value, the product of the two This will generate a very strong positive pulse. In normal load fluctuations, current increases and decreases typically do not reach their peak values simultaneously, or one of them may be negative, such as during acceleration. If it is negative, then... The value is much smaller than in the case of stalling. For example, if stalling occurs... , ,but During a normal load increase, It could be 0.05. It might be version 2.0, which yields... The difference is only 0.1, a significant difference in magnitude. As an alternative embodiment, a weighted summation method can also be used: ,in and The weights are positive constants, and the optimal signal-to-noise ratio is obtained through experimental calibration.
[0068] In a preferred embodiment, the method further includes:
[0069] Under stable operation of the micro motor, electromechanical deviation data for N sampling periods are continuously collected to form a sliding window; the moving average and moving standard deviation of the data within the sliding window are calculated in real time; the moving average and the moving standard deviation are added together and the sum is used as the dynamic preset threshold.
[0070] To achieve adaptive operation under different operating conditions, a dynamic threshold based on statistical process control is adopted. A first-in-first-out (FIFO) sliding window of length N is set to store the electromechanical deviation of the most recent N sampling points. Data. The selection of the window length N requires a trade-off between response speed and statistical stability; the preferred range is N = 200 to 1000. For example, for a 10kHz sampling rate, N = 500 is chosen, corresponding to a 50ms time window. At each sampling moment, when the new deviation... The calculated value is then added to the window, while the oldest data is removed. The moving average of 500 data points within the window is calculated in real-time. and moving standard deviation Dynamic threshold Based on the 3-sigma criterion, the specific calculation formula is as follows: The coefficient k is an adjustable sensitivity parameter with a value greater than or equal to 3, preferably 4 or 5, to effectively avoid misjudgments under drastic but normal load changes while ensuring sensitivity. For example, under a certain stable operating condition, the calculated... , If k=4 is selected, the current dynamic threshold is... To prevent false triggering caused by a single noise spike, triggering should only occur when a noise spike is detected. When the situation occurs continuously for M sampling points, M is a preset continuous counter, for example, M=50, that is, it lasts for 5ms before the stall event is confirmed and the protection action in S5 is immediately executed.
[0071] The second embodiment is a micromotor stall protection system based on current waveform, including the following modules:
[0072] The acquisition module is used to acquire the current signal of the stator winding of the micro motor as raw data during operation;
[0073] The decomposition module is used to perform variational mode decomposition on the original data, decomposing it into intrinsic mode functions (IMFs) of a preset target number of K modes; based on the center frequency of each IMF component, the fundamental current IMF component representing the power grid fundamental frequency, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics are selected.
[0074] The calculation module is used to perform a Hilbert transform on the fundamental current IMF component, extract its amplitude envelope, and differentiate it with respect to time to obtain the rate of change of power injection representing the instantaneous input power change; to perform a Hilbert transform on the specific high-frequency IMF component, extract its instantaneous frequency, differentiate it with respect to time, and then invert it to obtain the virtual mechanical deceleration representing the change of motor speed; and to combine the rate of change of power injection and the virtual mechanical deceleration to obtain the electromechanical deviation degree.
[0075] The comparison module is used to compare the currently calculated electromechanical deviation with a dynamic preset threshold. If the electromechanical deviation continues to exceed the dynamic preset threshold, it is determined that the micro motor is stalled.
[0076] The protection module is used to immediately block the PWM drive signal of the drive bridge and control the drive bridge to enter the active freewheeling mode after a stall occurs, so as to quickly dissipate the inductance energy stored in the stator winding and protect the micro motor.
[0077] In a preferred embodiment, the preset target mode number K is an optimal integer value that can effectively separate the fundamental current from specific high-frequency harmonics, determined in advance based on the electrical characteristics and mechanical structure of the micro-motor under test through offline experiments or simulation analysis.
[0078] In a preferred embodiment, the step of filtering out the fundamental current IMF component representing the grid fundamental frequency and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics, based on the center frequency of each IMF component, includes:
[0079] Calculate the center frequencies of K intrinsic mode functions (IMFs); take the IMF component whose center frequency is closest to the power supply frequency of the micromotor as the fundamental current IMF component; take one or more IMF components with high energy in a specific harmonic frequency band as specific high-frequency IMF components.
[0080] In a preferred embodiment, performing a Hilbert transform on the fundamental current IMF component, extracting its amplitude envelope, and differentiating it with respect to time to obtain the rate of change of power injection representing the instantaneous change in input power includes:
[0081] The fundamental current IMF component is subjected to Hilbert transform to obtain an analytical signal; the magnitude of the analytical signal is calculated to obtain the transient amplitude envelope; the rate of change of electrical energy injection is obtained by calculating the difference between the transient amplitude envelope at the current sampling time and the previous sampling time.
[0082] In a preferred embodiment, the step of performing a Hilbert transform on the specific high-frequency IMF component, extracting its instantaneous frequency, and then inverting the derivative with respect to time to obtain the virtual mechanical deceleration representing the change in motor speed includes:
[0083] If multiple specific high-frequency IMF components are selected, they are superimposed into a composite high-frequency signal; Hilbert transform is performed on the specific high-frequency IMF components or the composite high-frequency signal to obtain an analytical signal; the phase of the analytical signal is calculated and unwrapped; the unwrapped phase is subjected to first-order difference to obtain the instantaneous frequency; the instantaneous frequency is then subjected to first-order difference, and the result is inverted to obtain the virtual mechanical deceleration.
[0084] In a preferred embodiment, the step of combining the rate of change of electrical energy injection with the virtual mechanical deceleration to obtain the electromechanical deviation includes:
[0085] The electromechanical deviation is obtained by multiplying the rate of change of electrical energy injection by the virtual mechanical deceleration.
[0086] In a preferred embodiment, the method further includes:
[0087] Under stable operation of the micro motor, electromechanical deviation data for N sampling periods are continuously collected to form a sliding window; the moving average and moving standard deviation of the data within the sliding window are calculated in real time; the moving average and the moving standard deviation are added together and the sum is used as the dynamic preset threshold.
[0088] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.
[0089] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for stall protection of a micromotor based on current waveform, characterized in that, Includes the following steps: S1, acquire the current signal of the stator winding during micro-motor operation as raw data; S2, perform variational mode decomposition on the original data to decompose it into intrinsic mode functions (IMFs) of a preset target number of K modes; based on the center frequency of each IMF component, select the fundamental current IMF component representing the power grid fundamental frequency, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics. S3. Perform a Hilbert transform on the fundamental current IMF component, extract its amplitude envelope, and differentiate it with respect to time to obtain the rate of change of electrical energy injection representing the instantaneous input power change; perform a Hilbert transform on the specific high-frequency IMF component, extract its instantaneous frequency, differentiate it with respect to time, and then invert it to obtain the virtual mechanical deceleration representing the change of motor speed; combine the rate of change of electrical energy injection and the virtual mechanical deceleration to obtain the electromechanical deviation degree; S4. Compare the currently calculated electromechanical deviation with the dynamic preset threshold. If the electromechanical deviation continues to exceed the dynamic preset threshold, it is determined that the micro motor is stalled. S5, upon determining that a stall has occurred, immediately blocks the PWM drive signal of the drive bridge and controls the drive bridge to enter the active freewheeling mode to quickly dissipate the inductance energy stored in the stator winding, thereby protecting the micro motor.
2. The method according to claim 1, characterized in that, The preset target mode number K is an optimal integer value that can effectively separate the fundamental current from specific high-frequency harmonics, determined in advance based on the electrical characteristics and mechanical structure of the micro-motor under test through offline experiments or simulation analysis.
3. The method according to claim 1, characterized in that, The process of selecting the fundamental current IMF component representing the grid fundamental frequency based on the center frequency of each IMF component, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics, includes: Calculate the center frequencies of K intrinsic mode functions (IMFs); take the IMF component whose center frequency is closest to the power supply frequency of the micromotor as the fundamental current IMF component; take one or more IMF components with high energy in a specific harmonic frequency band as specific high-frequency IMF components.
4. The method according to claim 1, characterized in that, The step of performing a Hilbert transform on the fundamental current IMF component, extracting its amplitude envelope, and differentiating it with respect to time to obtain the rate of change of power injection representing the instantaneous change in input power includes: The fundamental current IMF component is subjected to Hilbert transform to obtain an analytical signal; the magnitude of the analytical signal is calculated to obtain the transient amplitude envelope; the rate of change of electrical energy injection is obtained by calculating the difference between the transient amplitude envelope at the current sampling time and the previous sampling time.
5. The method according to claim 1, characterized in that, The step of performing a Hilbert transform on the specific high-frequency IMF component, extracting its instantaneous frequency, and then inverting the derivative with respect to time to obtain the virtual mechanical deceleration representing the change in motor speed includes: If multiple specific high-frequency IMF components are selected, they are superimposed into a composite high-frequency signal; Hilbert transform is performed on the specific high-frequency IMF components or the composite high-frequency signal to obtain an analytical signal; the phase of the analytical signal is calculated and unwrapped; the unwrapped phase is subjected to first-order difference to obtain the instantaneous frequency; the instantaneous frequency is then subjected to first-order difference, and the result is inverted to obtain the virtual mechanical deceleration.
6. The method according to claim 1, characterized in that, The step of combining the rate of change of electrical energy injection with the virtual mechanical deceleration to obtain the electromechanical deviation includes: The electromechanical deviation is obtained by multiplying the rate of change of electrical energy injection by the virtual mechanical deceleration.
7. The method according to claim 1, characterized in that, The method further includes: Under stable operation of the micro motor, electromechanical deviation data for N sampling periods are continuously collected to form a sliding window; the moving average and moving standard deviation of the data within the sliding window are calculated in real time; the moving average and the moving standard deviation are added together and the sum is used as the dynamic preset threshold.
8. A micromotor stall protection system based on current waveform, characterized in that, Includes the following modules: The acquisition module is used to acquire the current signal of the stator winding of the micro motor as raw data during operation; The decomposition module is used to perform variational mode decomposition on the original data, decomposing it into intrinsic mode functions (IMFs) of a preset target number of K modes; based on the center frequency of each IMF component, the fundamental current IMF component representing the power grid fundamental frequency, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics are selected. The calculation module is used to perform a Hilbert transform on the fundamental current IMF component, extract its amplitude envelope, and differentiate it with respect to time to obtain the rate of change of power injection representing the instantaneous input power change; to perform a Hilbert transform on the specific high-frequency IMF component, extract its instantaneous frequency, differentiate it with respect to time, and then invert it to obtain the virtual mechanical deceleration representing the change of motor speed; and to combine the rate of change of power injection and the virtual mechanical deceleration to obtain the electromechanical deviation degree. The comparison module is used to compare the currently calculated electromechanical deviation with a dynamic preset threshold. If the electromechanical deviation continues to exceed the dynamic preset threshold, it is determined that the micro motor is stalled. The protection module is used to immediately block the PWM drive signal of the drive bridge and control the drive bridge to enter the active freewheeling mode after a stall occurs, so as to quickly dissipate the inductance energy stored in the stator winding and protect the micro motor.
9. The system according to claim 8, characterized in that, The preset target mode number K is an optimal integer value that can effectively separate the fundamental current from specific high-frequency harmonics, determined in advance based on the electrical characteristics and mechanical structure of the micro-motor under test through offline experiments or simulation analysis.
10. The system according to claim 8, characterized in that, The process of selecting the fundamental current IMF component representing the grid fundamental frequency based on the center frequency of each IMF component, and at least one specific high-frequency IMF component whose center frequency is related to the motor speed or structural harmonics, includes: Calculate the center frequencies of K intrinsic mode functions (IMFs); take the IMF component whose center frequency is closest to the power supply frequency of the micromotor as the fundamental current IMF component; take one or more IMF components with high energy in a specific harmonic frequency band as specific high-frequency IMF components.