Double three-phase permanent magnet motor system fault diagnosis method and system based on high-frequency signal injection

Abstract

The invention discloses a dual three-phase permanent magnet motor system fault diagnosis method and system based on high-frequency signal injection, and the method comprises the steps: accurately positioning a fault through analyzing the relation among a fundamental component, a harmonic component and a zero-sequence component before and after the fault; then, high-frequency signal injection is adopted to distinguish an open circuit from a current sensor fault; and finally, the fault position and type are judged and output. In the above process, a least square method based on a forgetting factor is designed to pre-process a fault positioning index, and a frequency tracking algorithm is used to extract an amplitude of a high-frequency current response to determine a fault type. According to the method, the problems of diagnosis and identification of the open-circuit fault of the dual three-phase permanent magnet synchronous motor and the fault of the current sensor are solved, fault diagnosis is carried out only by using the existing current signal in the control system, and additional hardware does not need to be installed. The method is simple and has very high robustness, and possibility is provided for improving the hardware utilization rate and fault-tolerant performance after a fault occurs.

Description

Technical Field

[0001] The present invention belongs to the technical field of motor fault diagnosis, and particularly relates to the diagnosis and identification of open - circuit faults and current sensor faults in a dual - three - phase permanent magnet motor drive system. Background Art

[0002] Due to its excellent fault - tolerance ability, the dual - three - phase permanent magnet synchronous motor has received increasing attention and applications in industrial applications such as electric vehicles, aerospace, and ship propulsion systems. With the continuous improvement of requirements for safety and reliability, the fault diagnosis technology of multi - phase permanent magnet motor drive systems has important engineering application value and theoretical research significance.

[0003] Due to reasons such as long - term operation under harsh conditions and component aging, open - circuit faults and current sensor faults are almost inevitable. In a multi - phase system, the probability of faults increases with the increase in the number of phases and current sensors. Although these two types of faults usually do not cause the motor drive system to shut down immediately, they will cause the sampled current value of the faulty phase to drop rapidly to zero, resulting in current imbalance, torque ripple, and even possible secondary faults. In addition, due to the similarity of fault characteristics, misdiagnosis is very likely to occur.

[0004] Domestic and foreign scholars have achieved certain results in the research on fault diagnosis of multi - phase motors. For the current sensor fault diagnosis method, currently, it mainly uses the design of corresponding observers to achieve, but it depends strongly on the accuracy of control system parameters. For open - circuit fault diagnosis methods, most are implemented by using current signal processing methods, such as the phase - current average value method, harmonic vector trajectory method, etc. By analyzing and processing the signals collected by current sensors, no additional sensors need to be added, which is easy to implement. However, since the signals obtained by current sensors are used for diagnosis, if the current sensor fails, it will have an impact on misdiagnosis. And the open - circuit fault characteristics are similar to the current sensor fault characteristics, and there is a coupling relationship between the two, which poses certain challenges to the robustness and accuracy of existing fault diagnosis methods.

[0005] Generally, open - circuit faults and current sensor faults are studied independently, which leads to a high risk of misdiagnosis, especially for current - signal - based methods. Therefore, it is necessary to conduct comprehensive diagnosis of open - circuit faults and current sensor faults. Existing comprehensive diagnosis methods require the installation of additional high - frequency voltage sensors in the control system, which increases the system cost and fault risk. Summary of the Invention

[0006] Aiming at the deficiencies in the existing technology, the present invention provides a fault diagnosis method for a dual - three - phase permanent magnet motor system based on high - frequency signal injection. Considering the influence of zero - sequence current, fault characteristics are constructed through the component currents of phase currents, and combined with the high - frequency injection method, the positioning and identification of open - circuit faults and current sensor faults are realized.

[0007] The present invention achieves the above technical objectives through the following technical means.

[0008] A fault diagnosis method for a dual-three-phase permanent magnet motor system based on high-frequency signal injection:

[0009] Collect the six-phase stator current of the dual-three-phase permanent magnet motor in real time, and transform it to the stationary coordinate system through coordinate transformation to obtain the fundamental current, harmonic current, and zero-sequence current, and further decompose them into fundamental component F k , harmonic component H k and zero-sequence component Z k ;

[0010] Extract fault features using the fundamental component, harmonic component, and zero-sequence component of each phase. According to the relationship of each component before and after the fault, define the fault location index, and preprocess the fault location index using the least squares method based on the forgetting factor;

[0011] Calculate the fault location flag bit according to the preprocessed fault location index and output the fault location;

[0012] Inject a high-frequency voltage signal according to the fault location, extract the amplitude of the high-frequency current response, determine the fault type and output.

[0013] Furthermore, extracting fault features using the fundamental component, harmonic component, and zero-sequence component of each phase specifically means that when the system is operating normally, H k +Z k =0; when an open circuit fault or current sensor fault occurs in phase k, H k +Z k =-F k ; when other phase faults occur, H k +Z k ≠-F k .

[0014] Even further, the fault location index When the system is operating normally, d k =0; when an open circuit fault or current sensor fault occurs in phase k, d k =1; when other phase faults occur, d k ≠1.

[0015] Even further, preprocess the fault location index using the least squares method based on the forgetting factor, specifically by performing parameter identification on the fault location index:

[0016]

[0017] In the formula, x i is the input data, corresponding to Fk ; y i is the output data, corresponding to H k +Z k ; P i is the covariance matrix, K i is the gain matrix, d k is the parameter to be identified, λ is the forgetting factor, and the subscript i is the number of beats.

[0018] Furthermore, calculate the fault location flag bit according to the fault location index and output the fault location, specifically:

[0019] When d k ∈(1 - ε, 1 + ε), the intermediate variable flag bit m corresponding to k k is set to 1, otherwise the intermediate variable flag bit corresponding to k is 0; perform a moving average filtering process on the intermediate variable flag bits of each phase and compare with the determination threshold M th Make a comparison. When <m k >> M th Let the fault flag bit D k = 1, otherwise D k = 0; where ε is the hysteresis bandwidth;

[0020] According to the fault location flag bit D k , output the fault location

[0021] Furthermore, inject a high-frequency voltage signal from the fault location, specifically:

[0022] Assume that the k-phase in the same set of windings is the fault phase, and the N1-phase and N2-phase are normal phases. The high-frequency voltage signal is defined as u kh = -2u N1h = -2u N1h = U h cos(ω h t); where U h , ω h are the amplitude and frequency of the injected high-frequency voltage signal respectively;

[0023] Then modulate the high-frequency voltage signal through a carrier to obtain the high-frequency signal duty cycle and inject it into the dual three-phase permanent magnet motor.

[0024] Even further, determine the fault type according to the amplitude of the high-frequency current response: If an open circuit fault occurs in the k-phase, no high-frequency injection voltage signal generates a high-frequency current signal response in the adjacent phases within the same winding, that is, the amplitude I of the high-frequency current response kh = 0; If a current sensor fault occurs in the k-phase, the adjacent phases within the same winding contain a high-frequency injection voltage signal generating a high-frequency current signal response, that is, I kh>I th , where I th is the threshold value of the high-frequency current response amplitude.

[0025] Furthermore, the output fault types are:

[0026] Furthermore, the coordinate transformation adopts the vector space decoupling method.

[0027] A fault diagnosis system for a dual-three-phase permanent magnet motor system based on high-frequency signal injection, comprising:

[0028] A current sampling and coordinate transformation module, configured to collect the six-phase stator currents of the dual-three-phase permanent magnet motor, and through coordinate transformation conversion and decomposition, obtain the fundamental wave component, harmonic component, and zero-sequence component of each phase current;

[0029] A fault location module, configured to define a fault location index, and preprocess the fault location index by using the least squares method based on the forgetting factor, calculate the fault location flag bit, and determine the fault location;

[0030] A fault type identification and output module, which injects a high-frequency voltage signal according to the fault location, extracts the amplitude of the high-frequency current response, determines the fault type, and outputs it.

[0031] The beneficial effects of the present invention are:

[0032] 1) The present invention can diagnose and identify open-circuit faults and current sensor faults in a dual-three-phase motor drive system, has a lower misdiagnosis probability and better practical applicability, is simple and efficient, and provides the possibility for adopting appropriate fault-tolerant strategies and greatly improving the hardware utilization rate.

[0033] 2) The diagnostic method proposed by the present invention uses the existing control signals in the system for diagnosis without the need to install additional hardware.

[0034] 3) In the drive system control of the present invention, the harmonics adopt closed-loop control. Among them, the defined fault location index, when the drive system is operating normally, the fundamental wave current is not zero, and the harmonic current and zero-sequence current are constrained to be zero. Therefore, the fault location index is always approximately zero during normal operation. Even when the speed / load changes suddenly, false alarms will not occur, and it has strong robustness. In addition, due to the strong constraint conditions of the vector space decoupling transformation, even under working conditions such as unbalanced power supply and low carrier ratio, when the harmonics are large, this method still has strong robustness.

[0035] 4) The fault diagnosis method proposed by the present invention, the diagnosis speed largely depends on the extraction speed of the high-frequency current response, and is weakly related to the fundamental wave frequency. Therefore, this method has great advantages at low speeds.

[0036] 5) The fault diagnosis technology proposed by the present invention is applicable to various polyphase motors, and is easy to implement, which is conducive to the engineering and practical application of new theories. Description of the Drawings

[0037] Figure 1 is the flowchart of the fault diagnosis method of the present invention;

[0038] Figure 2 is the hardware circuit structure diagram of the dual three-phase permanent magnet synchronous motor drive system of the present invention;

[0039] Figure 3 is the fault diagnosis diagram of the DTP-PMSM drive system based on FOC of the present invention;

[0040] Figure 4 is the flowchart of the frequency tracking algorithm adopted by the present invention;

[0041] Fig. 5(a) is the experimental result diagram of the robustness verification of the method of the present invention under the condition of torque mutation;

[0042] Fig. 5(b) is the experimental result diagram of the robustness verification of the method of the present invention under the condition of speed mutation;

[0043] Fig. 6(a) is the experimental result diagram of the open circuit fault diagnosis of phase A;

[0044] Fig. 6(b) is the experimental result diagram of the open circuit fault diagnosis of phase F;

[0045] Fig. 7(a) is the experimental result diagram of the fault diagnosis of the current sensor of phase A;

[0046] Fig. 7(b) is the experimental result diagram of the fault diagnosis of the current sensor of phase F. Detailed Embodiment

[0047] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0048] As Figure 1 shown is the flowchart of a fault diagnosis method for a dual three-phase permanent magnet motor system based on high-frequency signal injection according to the invention, which is applicable to the positioning and identification of open circuit faults and current sensor disconnection faults of dual three-phase permanent magnet motors, and can diagnose and output fault diagnosis results in real time.

[0049] Figure 2 shown is the schematic diagram of the hardware circuit structure in the normal state and fault state of the motor. Figure 3 shown is the application of the method of the present invention to the fault diagnosis of a dual three-phase permanent magnet synchronous motor drive system based on field-oriented control.

[0050] The method of the present invention specifically includes current sampling and coordinate transformation, fault location, fault type identification, and output of fault diagnosis results. The specific implementation steps are as follows (taking the fault of phase A of a dual-three-phase permanent magnet synchronous motor as an example):

[0051] Step 1) Use a current sensor to collect the six-phase stator currents of the dual-three-phase permanent magnet motor in real time, and transform them to the stationary coordinate system through vector space decoupling coordinate transformation to obtain fundamental current, harmonic current, and zero-sequence current. Decompose the stator current of each phase of the motor into fundamental component, harmonic component, and zero-sequence component independently;

[0052] Step 1.1) Use a current sensor to collect the six-phase stator currents (i A , i B , i C , i D , i E , i F ) in real time, and transform them to the six-phase stationary coordinate system through vector space decoupling coordinate transformation to obtain fundamental current (i α , i β ), harmonic current (i x , i y ), and zero-sequence current (i o1 , i o2 ):

[0053]

[0054] Step 1.2) Decompose the stator current of each phase of the motor into fundamental component F k , harmonic component H k , and zero-sequence component Z k ;

[0055] The fundamental components of phases A, B,..., F can be expressed as:

[0056]

[0057] The harmonic components of phases A, B,..., F can be expressed as:

[0058]

[0059] The zero-sequence components of phases A, B,..., F can be expressed as:

[0060]

[0061] The stator current of each phase can be uniformly expressed as i k = F k + H k + Z k , where k ∈ (A, B, C, D, E, F).

[0062] Step 2) Extract fault features using the fundamental wave components, harmonic components, and zero-sequence components of each phase, and design a fault location index d k , and perform preprocessing using the least squares method based on the forgetting factor;

[0063] Step 2.1) Extract fault features using the fundamental wave components, harmonic components, and zero-sequence components of each phase:

[0064] When the system is operating normally, the harmonic component H of each phase k and the zero-sequence component Z k are both 0, and F k = i k ; when an open-circuit fault occurs in the k-th phase, i k = 0, H k is a fluctuating value, and Z k = 0; when a current sensor fault occurs in the k-th phase, i k = 0, H k and Z k are both fluctuating values. Therefore, when the system is normal, H k + Z k = 0; when an open-circuit fault or a current sensor fault occurs in the k-th phase, H k + Z k = -F k ; when a fault occurs in other phases, H k + Z k ≠ -F k .

[0065] Step 2.2) Define the fault location index d according to the relationship between the component faults before and after k :

[0066]

[0067] When the system is normal, d k = 0; when an open-circuit fault or a current sensor fault occurs in the k-th phase, d k = 1; when a fault occurs in other phases, d k ≠ 1.

[0068] Step 2.3) Perform preprocessing on d k using the least squares method based on the forgetting factor:

[0069] Use the least squares method with forgetting factor iteration to perform parameter identification on the fault location index d k :

[0070]

[0071] where the subscript i is the number of beats; x iis the input data, corresponding to F k , y i is the output data, corresponding to H k +Z k ; K i is the gain matrix, d k is the parameter to be identified, P i is the covariance matrix, λ is the forgetting factor, and the forgetting factor is generally taken as 0.8 - 1, which can not only ensure the diagnosis speed but also have a good filtering effect.

[0072] Step 3) According to the preprocessed fault location index d k , calculate the fault location flag D k , and output the fault location F L ;

[0073] Step 3.1) According to the preprocessed fault location index d k , calculate the fault location flag D k , to achieve fault location;

[0074] When d k ∈(1 - ε, 1 + ε), the intermediate variable flag m k corresponding to k is set to 1, otherwise the corresponding intermediate variable flag is 0; where ε is the hysteresis bandwidth. Perform a moving average filtering process on the intermediate variable flags of each phase and compare with the decision threshold M th :

[0075]

[0076] When <m k >> M th , let the fault flag D k = 1, otherwise D k = 0;

[0077] Among them, T σ represents the moving average window length;

[0078] Step 3.2) According to the fault location flag D k , output the fault location F L ;

[0079]

[0080] Step 4) Inject a high-frequency voltage signal according to the fault location, extract the amplitude of the high-frequency current response through the frequency tracking algorithm, and determine and output the fault type F T ;

[0081] Step 4.1) According to the fault location F LDetermine the high-frequency injection voltage signal:

[0082] Assume that the k-phase is the faulty phase and the N1-phase and N2-phase are normal phases within the same set of windings. The high-frequency voltage signal can be defined as u kh =-2u N1h =-2u N1h =U h cos(ω h t); where U h and ω h are the amplitude and frequency of the injected high-frequency voltage signal respectively. When an open-circuit fault occurs in the A-phase, the high-frequency injection voltage signal is defined as:

[0083]

[0084] Then, modulate the high-frequency voltage signal through carrier to obtain the high-frequency signal duty cycle, and then inject it into the dual three-phase permanent magnet motor.

[0085] Step 4.2) Extract the amplitude of the high-frequency current response through the frequency tracking algorithm, as Figure 4 shown in the figure. In the figure, i kq and i kd are the components of the k-phase current on the d-axis and q-axis respectively. After being processed by the low-pass filter, i kql and i kdl are obtained. I kh is the amplitude of the high-frequency current response extracted from the k-phase.

[0086] Step 4.3) Determine the fault type according to the amplitude of the high-frequency current response:

[0087] If an open-circuit fault occurs in the k-phase, no high-frequency injection voltage signal in the adjacent phases within the same winding generates a high-frequency current signal response, that is, the amplitude of the high-frequency current response I kh =0; if a current sensor fault occurs in the k-phase, the adjacent phases within the same winding contain a high-frequency injection voltage signal generating a high-frequency current signal response, that is, I kh >I th , where I th is the threshold of the amplitude of the high-frequency current response;

[0088] Step 4.4) Output the fault type:

[0089]

[0090] When the system is normal, F T =0; after a fault occurs, if F T =1, then an open-circuit fault occurs in the k-phase of the system; if F T =2, then a current sensor fault occurs in the k-phase of the system. The specific fault type output results can be according to Table 1.

[0091] Table 1 Fault Type Table

[0092]

[0093] Next, verify the robustness of the fault diagnosis method proposed by the present invention under normal operation. Figure 5(a) shows the experimental result diagram of the robustness verification of the method of the present invention under the condition of torque mutation. The dual-three-phase permanent magnet synchronous motor runs at a speed of 100 r / min, the load torque changes from no-load to 160 N·m, the current of phase A increases from 0.5 A to 5.5 A, and the fault location flag F L and the fault identification flag F T are both zero. In Figure 5(b), under a load of 160 N·m, the test motor is driven from 100 revolutions per minute to 200 revolutions per minute. The fault location flag F L and the fault identification flag F T are both zero before and after the speed transient. The experimental results show that the fault diagnosis method proposed by the present invention has good robustness under speed step and load step.

[0094] Next, verify the effectiveness of the fault diagnosis proposed by the present invention in the case of open-circuit faults. As shown in Figures 6(a) and (b), the dual-three-phase permanent magnet synchronous motor operates at 100 r / min and 160 N·m, and the diagnostic results of open-circuit faults occurring in different phases. In the case of open-circuit faults, the sampled current and the actual current of the faulty phase are both zero. As shown in Figure 6(a), the fault location flag F L changes from 0 to 1, indicating that a fault occurs in phase A. The fault identification flag F T changes from 0 to 1, indicating that the fault type is an open-circuit fault. In Figure 6(b), the fault location flag F L changes from 0 to 6, and then the fault identification flag F T changes from 0 to 1, indicating that an open-circuit fault occurs in phase F. The results show that the method of the present invention has a certain ability to locate and identify open-phase faults occurring in different phases. In the case of open-circuit faults, the diagnostic times for fault location and identification are 9.55 ms and 32.65 ms respectively, accounting for 17.51% and 59.86% of the basic period respectively; therefore, the total time for fault diagnosis is 42.2 ms.

[0095] The effectiveness of the fault diagnosis proposed by the present invention in the case of current sensor failure is verified below. When the dual three-phase permanent magnet synchronous motor operates at 100r / min and 160N.m, the diagnosis results of current sensor failures occurring in different phases are shown in Figures 7(a) and (b). Under the action of the current sensor failure, the sampled value of the fault phase current drops to zero; while the actual current amplitude of the fault phase increases, accompanied by distortion caused by faulty operation and high-frequency voltage signal injection. In Figure 7(a), when the current sensor failure occurs in phase A, the fault location mark F L From 0 to 1, the fault identification flag F T From 0 to 2. Similarly, in FIG7(B), the fault location flag F L From 0 to 6, fault identification flag F T Changing from 0 to 2 indicates that a current sensor fault occurs in phase F. The diagnostic time for fault location and identification under current sensor fault is 7.55ms and 31.75ms, corresponding to 13.84% and 58.21% fundamental period, respectively; therefore, the total time for fault diagnosis is 39.3ms. In general, the fault diagnosis method proposed in the present invention has the ability to quickly locate faults and accurately identify fault types. In addition, according to the experimental results, the diagnosis speed of current sensor faults is faster than that of open circuit faults.

[0096] In addition, according to the diagnosis results shown in Figures 6(a), (b) and 7(a), (b), the diagnosis time includes the fault location time and the fault type identification time, and the fault location time accounts for a small proportion. Therefore, the diagnosis speed depends largely on the extraction speed of the high-frequency current response and is weakly correlated with the fundamental frequency. Since the extraction speed of the high-frequency current response is independent of the fundamental frequency, this method has a greater advantage over the conventional method of using fundamental current signals for fault diagnosis at low speeds.

[0097] The above is only a description of the preferred embodiment of the present invention, and its purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. For those skilled in the art, other advantages and variations can be easily associated with the above embodiments. Therefore, the present invention is not limited to the above embodiments, which are only used as examples to provide a detailed and exemplary description of one form of the present invention. Therefore, all equivalent changes or modifications made according to the principles and design ideas disclosed by the present invention are within the protection scope of the present invention.

Claims

1. A dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection, characterized in that: The six-phase stator current of the dual three-phase permanent magnet motor is collected in real time and converted to a stationary coordinate system through coordinate transformation to obtain the fundamental current, harmonic current and zero-sequence current, which are further decomposed into the fundamental component F k , harmonic component H k With zero sequence component Z k ; The fault characteristics are extracted by using the fundamental component, harmonic component and zero-sequence component of each phase, a fault location index is defined according to the relationship between each component before and after the fault, and the fault location index is preprocessed by using the least square method based on the forgetting factor; Calculate the fault location flag according to the preprocessed fault location index and output the fault location; A high-frequency voltage signal is injected from the fault position, the amplitude of the high-frequency current response is extracted, and the fault type is determined and output.

2. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 1 is characterized in that: The fault characteristics are extracted by using the fundamental component, harmonic component and zero sequence component of each phase. Specifically, when the system is operating normally, H k +Z k =0; when an open circuit fault or current sensor fault occurs in phase k, H k +Z k =-F k ; When other phases fail, H k +Z k ≠-F k .

3. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 2 is characterized in that: The fault localization index When the system is operating normally, k =0; when an open circuit fault or current sensor fault occurs in phase k, d k =1; when other phases fail, d k ≠1.

4. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 3 is characterized in that: The fault location index is preprocessed using the least square method based on the forgetting factor, specifically, parameters of the fault location index are identified: In the formula, x i is the input data, corresponding to F k ;y i is the output data, corresponding to H k +Z k ;P i is the covariance matrix, K i is the gain matrix, d k is the parameter to be identified, λ is the forgetting factor, and subscript i is the number of beats.

5. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 1 is characterized in that: The fault location flag is calculated according to the fault location index, and the fault location is output, specifically: When k ∈(1-ε,1+ε), then the intermediate variable flag m corresponding to k k Set to 1, otherwise the intermediate variable flag corresponding to k is 0; perform sliding average filtering on the intermediate variable flag of each phase and compare it with the judgment threshold M th For comparison, when <m k >>M th When the fault flag is set to D k =1, otherwise D k =0; where ε is the hysteresis bandwidth; According to the fault location flag D k , output fault location 6. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 1 is characterized in that: A high-frequency voltage signal is injected from the fault location, specifically: Assume that phase k in the same set of windings is the fault phase, phases N1 and N2 are normal phases, and the high-frequency voltage signal is defined as u kh =-2u N1h =-2u N1h =U h cos(ω h t); where U h ,ω h are the amplitude and frequency of the injected high-frequency voltage signal respectively; Then the high-frequency voltage signal is modulated by the carrier to obtain the duty cycle of the high-frequency signal, which is then injected into the dual three-phase permanent magnet motor.

7. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 6 is characterized in that: The fault type is determined based on the amplitude of the high-frequency current response: if an open circuit fault occurs in phase k, there is no high-frequency injection voltage signal in the adjacent phases of the same winding to generate a high-frequency current signal response, that is, the amplitude of the high-frequency current response I kh = 0; if the current sensor of phase k fails, the adjacent phases in the same winding contain high-frequency injection voltage signals to generate high-frequency current signal responses, that is, I kh >I th , where I th is the threshold of the high-frequency current response amplitude.

8. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 7 is characterized in that: The output fault types are:

9. The dual three-phase permanent magnet motor system fault diagnosis method based on high-frequency signal injection according to claim 1 is characterized in that: The coordinate transformation conversion adopts a vector space decoupling method.

10. A system for implementing the dual three-phase permanent magnet motor system fault diagnosis method based on high frequency signal injection as described in any one of claims 1 to 9, characterized in that: include: The current sampling and coordinate transformation module is used to collect the six-phase stator current of the dual three-phase permanent magnet motor, and obtain the fundamental component, harmonic component and zero-sequence component of each phase current through coordinate transformation and decomposition; A fault location module is used to define a fault location index, and pre-process the fault location index using a least square method based on a forgetting factor, calculate a fault location flag, and determine a fault location; The fault type identification and output module injects high-frequency voltage signals according to the fault location, extracts the amplitude of the high-frequency current response, determines the fault type and outputs it.

Images