Method and system for open-phase detection of permanent magnet synchronous motor

Abstract

The invention provides a method and a system for open-phase detection of a permanent magnet synchronous motor. The method comprises the following steps: periodically collecting three-phase current signals; taking the absolute value of the three-phase current signal in the period; performing low-pass filtering on the absolute value of the three-phase current signal to obtain a three-phase current amplitude; judging that the phase of which the current amplitude is continuously lower than the amplitudes of the other two phases has an open-phase fault; the method is simple in structure, high in real-time performance and strong in robustness, does not need an additional sensor, does not need complex calculation such as coordinate transformation, has low requirements on the performance and memory of the single-chip microcomputer, and can be easily realized in the single-chip microcomputer.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control and protection, and in particular to a method and system for detecting phase loss in a permanent magnet synchronous motor. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in electric vehicles, industrial servos, wind power generation, and other fields due to their high efficiency and low losses. However, during operation, the motor may experience phase loss due to loose connections, cable damage, or drive failure. Running a motor with a phase loss can cause severe vibration, reduced efficiency, and overheating and burnout. Therefore, rapid online detection of phase loss faults is crucial for stable system operation. Prior art methods typically utilize methods such as the zero-sequence current method (see Chinese Patent Publication No. CN103840437A), coordinate transformation (see Chinese Patent Publication No. CN117879424A), or state observers. However, these methods suffer from complexity, high sensor dependency, high cost, and poor real-time performance. The coordinate transformation method requires multiplication and trigonometric calculations, making it difficult to apply to microcontrollers that do not support these functions. The observer method relies on external sensors and imposes high costs. The zero-sequence current method determines whether the zero-sequence current exceeds a threshold. The amplitude of the zero-sequence current will change under different operating conditions, so it is easy to misjudge, and the zero-sequence current method cannot specifically detect the missing phase. Summary of the Invention

[0003] The present invention solves the problems of the existing phase loss detection method being complex, highly dependent on sensors, high cost, and poor real-time performance, and proposes a method and system for phase loss detection of a permanent magnet synchronous motor. The method and system have a simple structure, high real-time performance, and strong robustness, do not require additional sensors, and because complex calculations such as coordinate transformation are not required, the requirements for microcontroller performance and memory are also low, and the system can be easily implemented in a microcontroller.

[0004] In order to achieve the above objectives, the following technical solutions are proposed: A method for detecting phase loss in a permanent magnet synchronous motor comprises the following steps: S1, periodically collects three-phase current signals; S2, takes the absolute value of the three-phase current signal within the cycle; S3, low-pass filtering the absolute values ​​of the three-phase current signals to obtain the three-phase current amplitudes; S4, determining that a phase failure occurs in the phase whose current amplitude is continuously lower than the amplitudes of the other two phases.

[0005] The present invention obtains a smooth signal by performing low-pass filtering on the absolute values ​​of the collected three-phase currents, and then compares the current amplitudes after filtering. When the current amplitude of a phase is continuously lower than the amplitudes of the other two phases, it is determined that a phase loss fault has occurred in that phase. The present invention does not rely on voltage sensors and coordinate transformations, and the algorithm is simple to implement, and its real-time performance and robustness are superior to those of the prior art. The present invention can not only determine a phase loss fault, but also determine the specific phase that is missing, and locate the fault more accurately, which is beneficial for later fault repair and troubleshooting. The method can perform online real-time detection and does not require coordinate transformation, has a small amount of calculation, and does not require additional sensors, so there is no need to increase additional costs. Since the present invention does not require additional sensors and does not require complex calculations such as coordinate transformation, the performance and memory requirements of the single-chip microcomputer are also low, so the present invention can be easily implemented in a single-chip microcomputer, is more practical, and has great engineering application value.

[0006] Preferably, the S4 specifically includes the following steps: S41, determine whether the current amplitude of a phase is smaller than that of the other two phases. If so, increase the number of abnormal cycles of the phase by 1 and proceed to S42. If not, return to S1; S42, determine whether the number of abnormal cycles is greater than the set value N. If so, output the phase as an abnormal phase. If not, return to S1.

[0007] The present invention performs detection for multiple cycles in order to avoid the influence of environmental noise and reduce the false alarm rate. The cycle threshold N can be determined according to the actual working conditions. If the environment is relatively noisy, N can be set slightly larger.

[0008] Preferably, the S41 uses a polling method to determine whether the current amplitude of a phase is smaller than that of the other two phases. The specific steps are as follows: S411, determine whether the current amplitude of phase A and the current amplitude of phase B are both greater than the current amplitude of phase C. If so, increase the number of abnormal cycles of phase C by 1 and proceed to S42. If not, proceed to S412; S412, determine whether the current amplitude of phase A and the current amplitude of phase C are both greater than the current amplitude of phase B. If so, increase the number of abnormal cycles of phase B by 1 and proceed to S42. If not, proceed to S413. S413, determine whether the B-phase current amplitude and the C-phase current amplitude are both greater than the A-phase current amplitude. If so, increase the number of abnormal cycles of phase A by 1 and proceed to S42. If not, return to S1.

[0009] Preferably, when comparing the current amplitudes of a certain phase with those of the other two phases, the other two phases are multiplied by a preset proportional threshold value P before comparison.

[0010] During motor operation, the three-phase currents are sinusoidal waves with a phase difference of 120 degrees. First, the three-phase currents are regularly sampled. The absolute values ​​of each phase are taken to eliminate the effects of positive and negative current cancellation. This filtered phase current becomes exclusively positive. The current is then smoothed through a low-pass filter. Some solutions require calculating the effective value of the current, which requires square root calculation and places high demands on the microcontroller. Here, the average value is obtained by low-pass filtering the absolute values ​​of the phase currents. Finally, the three-phase amplitudes are compared. If a phase falls below a preset proportional threshold relative to the other two phases for multiple cycles, a phase loss fault is detected, and that phase becomes the missing phase. The proportional threshold and duration threshold can be calibrated and adjusted based on actual operating conditions. This online detection method allows for timely fault detection, enabling the controller to initiate protective actions and avoid further losses. Other commonly used methods include detection after coordinate transformation, which increases the amount of calculation. The present invention simply and directly calculates the three-phase current coordinate domain with a simple algorithm, does not require a large amount of calculation and additional sensors, is easy to implement on ordinary single-chip microcomputers, has high real-time performance, low filtering delay, and can quickly detect faults; some methods compare sinusoidal signals, which will continuously switch between positive and negative, increasing the complexity of judgment, and the calibration parameters are prone to errors. The present invention has strong robustness, and the combination of absolute value and filtering effectively suppresses harmonics and transient interference. In addition, the parameters are all positive numbers, which are easy to calibrate, and the proportional threshold P and set value N can be flexibly adjusted according to actual working conditions.

[0011] Preferably, the value range of the ratio threshold P is greater than 0 and less than or equal to Pmax, and the process of obtaining Pmax is as follows: According to steps S1-S3, the current amplitudes of x phases are collected when the phase is missing, and the current amplitudes are sorted from large to small to obtain a current amplitude series, and the average of the first y current amplitudes of the current amplitude series is obtained to obtain Pmax.

[0012] If a phase is missing, the corresponding filtering result will be 0. Due to the influence of noise, the present invention sets a threshold. If the signal level of one phase is continuously lower than a preset proportional threshold P of the other two phases for >N control cycles, a phase loss fault is considered to have occurred. The present invention sets the proportional threshold P to mitigate the influence of environmental noise and reduce the false alarm rate. The proportional threshold P can be adjusted according to actual operating conditions; the smaller the P value, the lower the probability of false triggering.

[0013] Preferably, x is equal to 3y, and x is a multiple of 3. The maximum value at the head of the current amplitude series is averaged, and the obtained Pmax can reduce the false alarm rate.

[0014] A permanent magnet synchronous motor phase loss detection system adopts the above-mentioned permanent magnet synchronous motor phase loss detection method, including a main control module and a three-phase current signal acquisition module, the main control module is provided with an absolute value taking unit, a processing unit and a current amplitude judgment unit, the processing unit is electrically connected to the absolute value taking unit and the current amplitude judgment unit respectively, the absolute value taking unit is electrically connected to the three-phase current signal acquisition module, the processing unit is electrically connected to a low-pass filter, and the main control module is electrically connected to a data storage unit and a timing unit.

[0015] Preferably, the low-pass filter includes a first-order discrete filter, a second-order discrete filter, an IIR filter, or a FIR filter.

[0016] Preferably, the main control module is provided with a fault output unit, which is electrically connected to the external receiving module. The main control module of the present invention is provided with a fault output unit which is connected to the external receiving module in a communication mode such as Bluetooth or WiFi.

[0017] Preferably, the main control module is a single chip microcomputer.

[0018] The beneficial effects of the present invention are as follows: the present invention does not rely on voltage sensors and coordinate transformation, the algorithm is simple to implement, and the real-time and robustness are better than the existing technology. The present invention can not only determine the phase loss fault, but also determine the specific phase that is missing, and locate the fault more accurately, which is beneficial for later fault repair and troubleshooting. The present method can be detected online in real time and does not require coordinate transformation. The amount of calculation is small and no additional sensors are required, so there is no need to increase additional costs. Since the present invention does not require additional sensors and does not require complex calculations such as coordinate transformation, the requirements for single-chip microcomputer performance and memory are also low. Therefore, the present invention can be easily implemented in a single-chip microcomputer, is more in line with reality, and has great engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of the method of the present invention.

[0020] Figure 2 It is a polling flow chart of the present invention.

[0021] Figure 3 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0022] Example 1: This embodiment proposes a method for detecting phase loss in a permanent magnet synchronous motor. Figure 1 , including the following steps: S1, periodically collects the original values ​​ia, ib, ic of the three-phase current signals; S2, take the absolute values ​​of the three-phase current original values ​​ia, ib, ic respectively, and obtain the three-phase absolute values ​​abs_ia, abs_ib, abs_ic; S3, perform low-pass filtering on the three-phase current absolute values ​​abs_ia, abs_ib, abs_ic to obtain the filtered signals, namely the current amplitudes filter_abs_ia, filter_abs_ib, filter_abs_ic. The smaller the filter cutoff frequency, the more stable and smooth the obtained filter value; S4, compare filter_abs_ia, filter_abs_ib, filter_abs_ic. If a phase is missing, the corresponding filtering result will be 0, and it is determined that a phase loss fault occurs in the phase whose current amplitude is continuously lower than the amplitudes of the other two phases.

[0023] The S4 of the present invention specifically includes the following steps: S41, determine whether the current amplitude of a phase is smaller than that of the other two phases. If not, return to S1; if so, increase the number of abnormal cycles of the phase by 1 and proceed to S42; S42, determine whether the number of abnormal cycles is greater than the set value N, if not, return to S1; if so, output the phase as an abnormal phase.

[0024] The present invention performs detection for multiple cycles in order to avoid the influence of environmental noise and reduce the false alarm rate. The cycle threshold N can be determined according to the actual working conditions. If the environment is relatively noisy, N can be set slightly larger.

[0025] The present invention obtains a smooth signal by performing low-pass filtering on the absolute values ​​of the collected three-phase currents, and then compares the current amplitudes after filtering. When the current amplitude of a phase is continuously lower than the amplitudes of the other two phases, it is determined that a phase loss fault has occurred in that phase. The present invention does not rely on voltage sensors and coordinate transformations, and the algorithm is simple to implement, and its real-time performance and robustness are superior to those of the prior art. The present invention can not only determine a phase loss fault, but also determine the specific phase that is missing, and locate the fault more accurately, which is beneficial for later fault repair and troubleshooting. The method can perform online real-time detection and does not require coordinate transformation, has a small amount of calculation, and does not require additional sensors, so there is no need to increase additional costs. Since the present invention does not require additional sensors and does not require complex calculations such as coordinate transformation, the performance and memory requirements of the single-chip microcomputer are also low, so the present invention can be easily implemented in a single-chip microcomputer, is more practical, and has great engineering application value.

[0026] This embodiment also proposes a permanent magnet synchronous motor phase loss detection system, which adopts the above-mentioned permanent magnet synchronous motor phase loss detection method. Figure 3, including a main control module, namely a single-chip microcomputer, a three-phase current signal acquisition module and a low-pass filter. The main control module is provided with an absolute value acquisition unit, a processing unit, a current amplitude judgment unit, a data storage unit and a timing unit. The absolute value acquisition unit is electrically connected to the three-phase current signal acquisition module. The absolute value acquisition unit receives the three-phase current signal collected by the three-phase current signal acquisition module and converts it into the absolute value of the three-phase current signal. The absolute value acquisition unit is electrically connected to the processing unit and sends the extracted absolute value of the three-phase current signal to the processing unit. The processing unit is electrically connected to the low-pass filter. The processing unit sends the absolute value of the three-phase current signal to the low-pass filter. After processing by the low-pass filter, the current amplitude is fed back to the processing unit. The processing unit is electrically connected to the current amplitude judgment unit. The processing unit sends the current amplitude to the current amplitude judgment unit. The current amplitude judgment unit sends the judgment result to the processing unit. The data storage unit and the timing unit are respectively electrically connected to the processing unit. The data storage unit is used to store all collected and processed data. The timing unit is used to time the acquisition cycle. The processing unit of the present invention is provided with an execution program for implementing the above-mentioned permanent magnet synchronous motor phase loss detection method.

[0027] The low-pass filter of the present invention can be a first-order or second-order discrete filter or a filter of other forms such as an IIR or FIR filter. However, a first-order low-pass filter generally has the smallest computational complexity, so the best choice is a first-order low-pass filter.

[0028] Example 2: This embodiment improves S41 on the basis of embodiment 1 and proposes a method for detecting phase loss in a permanent magnet synchronous motor. Figure 1 , including the following steps: S1, periodically collects the original values ​​ia, ib, ic of the three-phase current signals; S2, take the absolute values ​​of the three-phase current original values ​​ia, ib, ic respectively, and obtain the three-phase absolute values ​​abs_ia, abs_ib, abs_ic; S3, perform low-pass filtering on the three-phase current absolute values ​​abs_ia, abs_ib, abs_ic to obtain the filtered signals, namely the current amplitudes filter_abs_ia, filter_abs_ib, filter_abs_ic. The smaller the filter cutoff frequency, the more stable and smooth the obtained filter value; S4, compare filter_abs_ia, filter_abs_ib, filter_abs_ic. If a phase is missing, the corresponding filtering result will be 0, and it is determined that a phase loss fault occurs in the phase whose current amplitude is continuously lower than the amplitudes of the other two phases.

[0029] The S4 of the present invention specifically includes the following steps: S41, determine whether the current amplitude of a phase is smaller than that of the other two phases. If not, return to S1; if so, increase the number of abnormal cycles of the phase by 1 and proceed to S42; S42, determine whether the number of abnormal cycles is greater than the set value N, if not, return to S1; if so, output the phase as an abnormal phase.

[0030] The present invention performs detection for multiple cycles in order to avoid the influence of environmental noise and reduce the false alarm rate. The cycle threshold N can be determined according to the actual working conditions. If the environment is relatively noisy, N can be set slightly larger.

[0031] The S41 of the present invention uses a polling method to determine whether there is a phase current amplitude smaller than the other two phases, referring to Figure 2 , the specific steps are as follows: S411, determine whether the A-phase current amplitude filter_abs_ia and the B-phase current amplitude filter_abs_ib are both greater than the C-phase current amplitude filter_abs_ic. If so, increase the number of abnormal cycles of phase C by 1 and proceed to S42. If not, proceed to S412. S412, determine whether the A-phase current amplitude filter_abs_ia and the C-phase current amplitude filter_abs_ic are both greater than the B-phase current amplitude filter_abs_ib. If so, increase the number of abnormal cycles of phase B by 1 and proceed to S42. If not, proceed to S413. S413, determine whether the B-phase current amplitude filter_abs_ib and the C-phase current amplitude filter_abs_ic are both greater than the A-phase current amplitude filter_abs_ia. If so, increase the number of abnormal cycles of phase A by 1 and proceed to S42. If not, return to S1.

[0032] The present invention obtains a smooth signal by performing low-pass filtering on the absolute values ​​of the collected three-phase currents, and then compares the current amplitudes after filtering. When the current amplitude of a phase is continuously lower than the amplitudes of the other two phases, it is determined that a phase loss fault has occurred in that phase. The present invention does not rely on voltage sensors and coordinate transformations, and the algorithm is simple to implement, and its real-time performance and robustness are superior to those of the prior art. The present invention can not only determine a phase loss fault, but also determine the specific phase that is missing, and locate the fault more accurately, which is beneficial for later fault repair and troubleshooting. The method can perform online real-time detection and does not require coordinate transformation, has a small amount of calculation, and does not require additional sensors, so there is no need to increase additional costs. Since the present invention does not require additional sensors and does not require complex calculations such as coordinate transformation, the performance and memory requirements of the single-chip microcomputer are also low, so the present invention can be easily implemented in a single-chip microcomputer, is more practical, and has great engineering application value.

[0033] This embodiment also proposes a permanent magnet synchronous motor phase loss detection system, which adopts the above-mentioned permanent magnet synchronous motor phase loss detection method. Figure 3 , including a main control module, namely a single-chip microcomputer, a three-phase current signal acquisition module and a low-pass filter. The main control module is provided with an absolute value acquisition unit, a processing unit, a current amplitude judgment unit, a data storage unit and a timing unit. The absolute value acquisition unit is electrically connected to the three-phase current signal acquisition module. The absolute value acquisition unit receives the three-phase current signal collected by the three-phase current signal acquisition module and converts it into the absolute value of the three-phase current signal. The absolute value acquisition unit is electrically connected to the processing unit and sends the extracted absolute value of the three-phase current signal to the processing unit. The processing unit is electrically connected to the low-pass filter. The processing unit sends the absolute value of the three-phase current signal to the low-pass filter. After processing by the low-pass filter, the current amplitude is fed back to the processing unit. The processing unit is electrically connected to the current amplitude judgment unit. The processing unit sends the current amplitude to the current amplitude judgment unit. The current amplitude judgment unit sends the judgment result to the processing unit. The data storage unit and the timing unit are respectively electrically connected to the processing unit. The data storage unit is used to store all collected and processed data. The timing unit is used to time the acquisition cycle. The processing unit of the present invention is provided with an execution program for implementing the above-mentioned permanent magnet synchronous motor phase loss detection method.

[0034] The low-pass filter of the present invention can be a first-order or second-order discrete filter or a filter of other forms such as an IIR or FIR filter. However, a first-order low-pass filter generally has the smallest computational complexity, so the best choice is a first-order low-pass filter.

[0035] Example 3: This embodiment further improves S41 on the basis of embodiment 1 and embodiment 2, and proposes a method for detecting phase loss in a permanent magnet synchronous motor. Figure 1 , including the following steps: S1, periodically collects the original values ​​ia, ib, ic of the three-phase current signals; S2, take the absolute values ​​of the three-phase current original values ​​ia, ib, ic respectively, and obtain the three-phase absolute values ​​abs_ia, abs_ib, abs_ic; S3, perform low-pass filtering on the three-phase current absolute values ​​abs_ia, abs_ib, abs_ic to obtain the filtered signals, namely the current amplitudes filter_abs_ia, filter_abs_ib, filter_abs_ic. The smaller the filter cutoff frequency, the more stable and smooth the obtained filter value; S4, compare filter_abs_ia, filter_abs_ib, filter_abs_ic. If a phase is missing, the corresponding filtering result will be 0, and it is determined that a phase loss fault occurs in the phase whose current amplitude is continuously lower than the amplitudes of the other two phases.

[0036] The S4 of the present invention specifically includes the following steps: S41, determine whether the current amplitude of a phase is smaller than that of the other two phases. If not, return to S1; if so, increase the number of abnormal cycles of the phase by 1 and proceed to S42; S42, determine whether the number of abnormal cycles is greater than the set value N, if not, return to S1; if so, output the phase as an abnormal phase.

[0037] The present invention performs detection for multiple cycles in order to avoid the influence of environmental noise and reduce the false alarm rate. The cycle threshold N can be determined according to the actual working conditions. If the environment is relatively noisy, N can be set slightly larger.

[0038] The S41 of the present invention uses a polling method to determine whether there is a phase current amplitude smaller than the other two phases, referring to Figure 2 , the specific steps are as follows: S411, determine whether the A-phase current amplitude filter_abs_ia and the B-phase current amplitude filter_abs_ib are both greater than the C-phase current amplitude filter_abs_ic. If so, increase the number of abnormal cycles of phase C by 1 and proceed to S42. If not, proceed to S412. S412, determine whether the A-phase current amplitude filter_abs_ia and the C-phase current amplitude filter_abs_ic are both greater than the B-phase current amplitude filter_abs_ib. If so, increase the number of abnormal cycles of phase B by 1 and proceed to S42. If not, proceed to S413. S413, determine whether the B-phase current amplitude filter_abs_ib and the C-phase current amplitude filter_abs_ic are both greater than the A-phase current amplitude filter_abs_ia. If so, increase the number of abnormal cycles of phase A by 1 and proceed to S42. If not, return to S1.

[0039] In S41 of the present invention, when comparing the current amplitude of a certain phase with that of the other two phases, the other two phases are multiplied by a preset proportional threshold P before comparison. For example, when judging phase C, it is determined whether the product of the current amplitude of phase A filter_abs_ia and the proportional threshold P and the product of the current amplitude of phase B filter_abs_ib and the proportional threshold P are both greater than the current amplitude of phase C filter_abs_ic; when judging phase B, it is determined whether the product of the current amplitude of phase A filter_abs_ia and the proportional threshold P and the product of the current amplitude of phase C filter_abs_ic and the proportional threshold P are both greater than the current amplitude of phase B filter_abs_ib; when judging phase A, it is determined whether the product of the current amplitude of phase C filter_abs_ic and the proportional threshold P and the product of the current amplitude of phase B filter_abs_ib and the proportional threshold P are both greater than the current amplitude of phase A filter_abs_ia.

[0040] During motor operation, the three-phase currents are sinusoidal waves with a phase difference of 120 degrees. First, the three-phase currents are regularly sampled. The absolute values ​​of each phase are taken to eliminate the effects of positive and negative current cancellation. This filtered phase current becomes exclusively positive. The current is then smoothed through a low-pass filter. Some solutions require calculating the effective value of the current, which requires square root calculation and places high demands on the microcontroller. Here, the average value is obtained by low-pass filtering the absolute values ​​of the phase currents. Finally, the three-phase amplitudes are compared. If a phase falls below a preset proportional threshold relative to the other two phases for multiple cycles, a phase loss fault is detected, and that phase becomes the missing phase. The proportional threshold and duration threshold can be calibrated and adjusted based on actual operating conditions. This online detection method allows for timely fault detection, enabling the controller to initiate protective actions and avoid further losses. Other commonly used methods include detection after coordinate transformation, which increases the amount of calculation. The present invention simply and directly calculates the three-phase current coordinate domain with a simple algorithm, does not require a large amount of calculation and additional sensors, is easy to implement on ordinary single-chip microcomputers, has high real-time performance, low filtering delay, and can quickly detect faults; some methods compare sinusoidal signals, which will continuously switch between positive and negative, increasing the complexity of judgment, and the calibration parameters are prone to errors. The present invention has strong robustness, and the combination of absolute value and filtering effectively suppresses harmonics and transient interference. In addition, the parameters are all positive numbers, which are easy to calibrate, and the proportional threshold P and set value N can be flexibly adjusted according to actual working conditions.

[0041] The minimum value of the ratio threshold P of the present invention is greater than 0, and the maximum value of the ratio threshold P is less than or equal to Pmax. The acquisition process of Pmax of the present invention is as follows: When each phase is missing, x current amplitudes are collected according to steps S1-S3, and the current amplitudes are sorted from large to small to obtain a current amplitude series, and the average of the first y current amplitudes of the current amplitude series is calculated, and the average is the final Pmax.

[0042] If a phase is missing, the corresponding filtering result will be 0. Due to the influence of noise, the present invention sets a threshold. If the signal level of one phase is continuously lower than a preset proportional threshold P of the other two phases for >N control cycles, a phase loss fault is considered to have occurred. The present invention sets the proportional threshold P to mitigate the influence of environmental noise and reduce the false alarm rate. The proportional threshold P can be adjusted according to actual operating conditions; the smaller the P value, the lower the probability of false triggering.

[0043] As a preferred solution of the present invention, x is set to a multiple of 3 when taking a value, and 3y is equal to x, that is, the maximum value at the head of the current amplitude series is averaged, and the obtained Pmax can reduce the false alarm rate.

[0044] The present invention obtains a smooth signal by performing low-pass filtering on the absolute values ​​of the collected three-phase currents, and then compares the current amplitudes after filtering. When the current amplitude of a phase is continuously lower than the amplitudes of the other two phases, it is determined that a phase loss fault has occurred in that phase. The present invention does not rely on voltage sensors and coordinate transformations, and the algorithm is simple to implement, and its real-time performance and robustness are superior to those of the prior art. The present invention can not only determine a phase loss fault, but also determine the specific phase that is missing, and locate the fault more accurately, which is beneficial for later fault repair and troubleshooting. The method can perform online real-time detection and does not require coordinate transformation, has a small amount of calculation, and does not require additional sensors, so there is no need to increase additional costs. Since the present invention does not require additional sensors and does not require complex calculations such as coordinate transformation, the performance and memory requirements of the single-chip microcomputer are also low, so the present invention can be easily implemented in a single-chip microcomputer, is more practical, and has great engineering application value.

[0045] This embodiment also proposes a permanent magnet synchronous motor phase loss detection system, which adopts the above-mentioned permanent magnet synchronous motor phase loss detection method. Figure 3, including a main control module, namely a single-chip microcomputer, a three-phase current signal acquisition module and a low-pass filter. The main control module is provided with an absolute value acquisition unit, a processing unit, a current amplitude judgment unit, a data storage unit and a timing unit. The absolute value acquisition unit is electrically connected to the three-phase current signal acquisition module. The absolute value acquisition unit receives the three-phase current signal collected by the three-phase current signal acquisition module and converts it into the absolute value of the three-phase current signal. The absolute value acquisition unit is electrically connected to the processing unit and sends the extracted absolute value of the three-phase current signal to the processing unit. The processing unit is electrically connected to the low-pass filter. The processing unit sends the absolute value of the three-phase current signal to the low-pass filter. After processing by the low-pass filter, the current amplitude is fed back to the processing unit. The processing unit is electrically connected to the current amplitude judgment unit. The processing unit sends the current amplitude to the current amplitude judgment unit. The current amplitude judgment unit sends the judgment result to the processing unit. The data storage unit and the timing unit are respectively electrically connected to the processing unit. The data storage unit is used to store all collected and processed data. The timing unit is used to time the acquisition cycle. The processing unit of the present invention is provided with an execution program for implementing the above-mentioned permanent magnet synchronous motor phase loss detection method.

[0046] The low-pass filter of the present invention can be a first-order or second-order discrete filter or a filter of other forms such as an IIR or FIR filter. However, a first-order low-pass filter generally has the smallest computational complexity, so the best choice is a first-order low-pass filter.

[0047] The main control module of the present invention is provided with a fault output unit that is connected to the external receiving module for communication, and the communication method can be Bluetooth or WiFi.

Claims

1. A method for detecting phase loss in a permanent magnet synchronous motor, characterized in that: The following steps are involved: S1, periodically collects three-phase current signals; S2, takes the absolute value of the three-phase current signal within the cycle; S3, low-pass filtering the absolute values ​​of the three-phase current signals to obtain the three-phase current amplitudes; S4, determining that a phase failure occurs in the phase whose current amplitude is continuously lower than the amplitudes of the other two phases.

2. The method for detecting phase loss in a permanent magnet synchronous motor according to claim 1, wherein: The S4 specifically includes the following steps: S41, determine whether the current amplitude of a phase is smaller than that of the other two phases. If so, increase the number of abnormal cycles of the phase by 1 and proceed to S42. If not, return to S1; S42, determine whether the number of abnormal cycles is greater than the set value N. If so, output the phase as an abnormal phase. If not, return to S1.

3. The method for detecting phase loss in a permanent magnet synchronous motor according to claim 2, wherein: The S41 uses a polling method to determine whether the current amplitude of a phase is smaller than that of the other two phases. The specific steps are as follows: S411, determine whether the current amplitude of phase A and the current amplitude of phase B are both greater than the current amplitude of phase C. If so, increase the number of abnormal cycles of phase C by 1 and proceed to S42. If not, proceed to S412; S412, determine whether the current amplitude of phase A and the current amplitude of phase C are both greater than the current amplitude of phase B. If so, increase the number of abnormal cycles of phase B by 1 and proceed to S42. If not, proceed to S413. S413, determine whether the B-phase current amplitude and the C-phase current amplitude are both greater than the A-phase current amplitude. If so, increase the number of abnormal cycles of phase A by 1 and proceed to S42. If not, return to S1.

4. The method for detecting phase loss in a permanent magnet synchronous motor according to claim 2, wherein When comparing the current amplitude of a phase with the other two phases, the other two phases are multiplied by a preset proportional threshold value P before comparison.

5. The method for detecting phase loss in a permanent magnet synchronous motor according to claim 4, wherein: The value range of the ratio threshold P is greater than 0 and less than or equal to Pmax. The process of obtaining Pmax is as follows: According to steps S1-S3, the current amplitudes of x phases are collected when the phase is missing, and the current amplitudes are sorted from large to small to obtain a current amplitude series, and the average of the first y current amplitudes of the current amplitude series is obtained to obtain Pmax.

6. The method for detecting phase loss in a permanent magnet synchronous motor according to claim 5, wherein: The x is equal to 3y, and x is a multiple of 3.

7. A permanent magnet synchronous motor phase loss detection system, using a permanent magnet synchronous motor phase loss detection method according to any one of claims 1 to 6, characterized in that: It includes a main control module and a three-phase current signal acquisition module. The main control module is provided with an absolute value taking unit, a processing unit and a current amplitude judgment unit. The processing unit is electrically connected to the absolute value taking unit and the current amplitude judgment unit respectively. The absolute value taking unit is electrically connected to the three-phase current signal acquisition module. The processing unit is electrically connected to a low-pass filter. The main control module is electrically connected to a data storage unit and a timing unit.

8. A permanent magnet synchronous motor phase loss detection system according to claim 7, characterized in that: The low-pass filter includes a first-order discrete filter, a second-order discrete filter, an IIR filter, or a FIR filter.

9. A permanent magnet synchronous motor phase loss detection system according to claim 7, characterized in that: The main control module is provided with a fault output unit, and the fault output unit is electrically connected to the external receiving module.

10. A permanent magnet synchronous motor phase loss detection system according to claim 8 or 9, characterized in that: The main control module is a single chip microcomputer.

Images