Permanent magnet synchronous motor fault diagnosis, location and classification method and system

Abstract

The present invention relates to the technical field of motor fault diagnosis and state monitoring, and specifically discloses a method and system for diagnosing, locating and classifying permanent magnet synchronous motor faults. The diagnostic method and system extract residual signals through a model predictive control (MPC) strategy, perform coordinate transformation on the current residual, extract its DC component, and use the absolute difference of the DC component as a fault diagnosis indicator to determine whether an inter-turn short circuit fault or a high resistance connection fault has occurred. The present invention can determine whether two types of faults, namely inter-turn short circuit (ITF) and high resistance connection (HRC) faults, have occurred, and further determine the location and type of fault. It has high diagnostic, positioning and classification accuracy and robustness, and can adapt to complex working conditions such as speed, load changes and parameter mismatches. In addition, the fault diagnosis and classification process can be implemented online in real time and efficiently, with simple operation and strong real-time performance.

Description

Technical Field

[0001] The present invention relates to the technical field of motor fault diagnosis and state monitoring, and in particular to a method and system for diagnosing, locating and classifying permanent magnet synchronous motor faults. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in industrial drives, transportation, and energy due to their high efficiency, reliability, and power density. However, as motors operate longer, stator windings can fail due to insulation aging, environmental factors, and mechanical stress. Inter-turn faults (ITFs) and high-resistance connections (HRCs) are the most common and potentially harmful types of failures. These faults can degrade motor performance and even cause serious equipment damage, impacting system safety and reliability.

[0003] There are significant differences in the formation mechanisms and manifestations of turn-to-turn short-circuit faults and high-resistance connection faults. Turn-to-turn short-circuit faults are typically caused by insulation damage and are characterized by an imbalance in stator current and a sharp increase in heat loss, which can quickly develop into a more serious phase-to-phase short-circuit or ground fault. High-resistance connection faults, on the other hand, manifest as increased resistance at the connection point, causing current fluctuations and voltage anomalies. Due to the distinct severity and handling methods of the two types of faults, fault classification is crucial in motor maintenance strategies. For example, turn-to-turn short-circuit faults require rapid detection and intervention to prevent further expansion, while high-resistance connection faults require precise analysis of the changing trends of the fault point to optimize maintenance timing.

[0004] Currently, fault detection methods based on multi-channel current signature analysis (MCSA) have become the mainstream technology for motor fault diagnosis due to their simplicity and ease of implementation. However, in closed-loop control systems, the controller actively reduces current imbalance by adjusting the voltage vector, making it difficult for MCSA methods to capture effective features in early fault detection, especially in the early stages of turn-to-turn short-circuit faults. High-resistance connection faults are even more difficult to quickly and accurately identify using traditional methods due to their weak and slowly changing characteristic signals. Furthermore, existing methods focus primarily on fault detection accuracy, with limited research on fault type classification, making it difficult to meet the demand for accurate classification of different faults in practical applications. Summary of the Invention

[0005] The present invention provides a method and system for diagnosing, locating and classifying permanent magnet synchronous motor faults, which solve the technical problems of difficulty in effectively capturing fault characteristics in a closed-loop control system, insufficient classification capability, and poor real-time and robustness.

[0006] To solve the above technical problems, the present invention provides a permanent magnet synchronous motor fault diagnosis method, comprising the steps of:

[0007] A1. Calculate the residual current of the α and β axes by the difference between the predicted current of the α and β axes and the actual measured current of the α and β axes of the three-phase permanent magnet synchronous motor;

[0008] A2. Squaring the residual currents of the α and β axes and extracting their DC components, and using the absolute difference of the DC components as a fault diagnosis indicator;

[0009] A3. In a fixed healthy state, 1.5 times the absolute difference between the DC components extracted after the α-axis and β-axis residual currents are multiplied by themselves is used as the fault diagnosis index threshold. The fault diagnosis index is compared with the fault diagnosis index threshold. If the fault diagnosis index is greater than the fault diagnosis index threshold, it is determined that a turn-to-turn short circuit fault or a high-resistance connection fault has occurred. Otherwise, it is determined to be normal.

[0010] The present invention also provides a permanent magnet synchronous motor fault diagnosis system, which comprises a residual current acquisition module, a fault diagnosis index calculation module and a fault diagnosis module, which are respectively used to execute steps A1 to A3 in the permanent magnet synchronous motor fault diagnosis method.

[0011] The present invention also provides a permanent magnet synchronous motor fault location method. Based on the permanent magnet synchronous motor fault diagnosis method, the key is that after step A3 determines that a turn-to-turn short circuit fault or a high resistance connection fault has occurred, the diagnostic method includes the following steps:

[0012] B1. Perform coordinate transformation on the α-axis and β-axis residual currents to rotate them to align with phases B and C, and obtain the residual currents in the coordinate system aligned with phases B and C.

[0013] B2. Squaring the rotated residual current with itself and extracting the DC component through low-pass filtering, and using it together with the DC components of the α-axis and β-axis residual currents as fault location features;

[0014] B3. Calculate the fault location index based on the fault location characteristics;

[0015] B4. Locate the fault according to the fault location indicators and determine the location where the fault occurred.

[0016] Furthermore, in step B3, the fault location indicators include:

[0017]

[0018] Among them, FI loc_A The residual current of α and β axes at time k is squared and the DC component is extracted by low-pass filtering. The difference, FI loc_BThe residual current of phase B after rotation at time k is squared and the DC component is extracted by low-pass filtering. The difference, FI loc_C The C-phase residual current after rotation at time k is squared and the DC component is extracted by low-pass filtering. difference

[0019] Furthermore, the step B4 is specifically as follows:

[0020] If FI loc_A >0 and FI loc_B >0 or FI loc_A <0 and FI loc_B <0, it is determined that the fault occurs in phase C;

[0021] If FI loc_A >0 and FI loc_C <0, it is determined that the fault occurs in phase A;

[0022] If FI loc_A <0 and FI loc_B >0, it is determined that the fault occurs in phase B.

[0023] The present invention also provides a permanent magnet synchronous motor fault location system, the key of which is that it includes a coordinate transformation module, a fault location feature calculation module, a fault location index calculation module, and a fault location module, which are respectively used to execute steps B1 to B4 in the permanent magnet synchronous motor fault diagnosis method.

[0024] The present invention also provides a permanent magnet synchronous motor fault classification method. Based on the permanent magnet synchronous motor fault diagnosis method, the key is that after step A3 determines that a turn-to-turn short circuit fault or a high resistance connection fault has occurred, the diagnosis method includes the following steps:

[0025] C1. Multiply the α-axis and β-axis residual currents with the fault phase current and extract their DC components through a low-pass filter to obtain classification features.

[0026] C2. Calculate fault classification index based on classification features;

[0027] C3. Classify the fault according to the fault classification index to determine whether the fault type is a turn-to-turn short circuit fault or a high-resistance connection fault.

[0028] Furthermore, in step C2, the fault classification index is the α and β axis classification features FI cla_α , FI cla_β The difference between FI cla =FI cla_α -FI cla_β .

[0029] Furthermore, the step C3 is specifically as follows: setting the fault classification threshold Thre cla =0, when FI cla >Thre cla If the fault is positive, it is determined as a high-resistance connection fault; otherwise, it is determined as an inter-turn short-circuit fault.

[0030] The present invention also provides a permanent magnet synchronous motor fault classification system, the key of which is that it includes a fault classification feature calculation module, a fault classification index calculation module, and a fault classification module, which are respectively used to execute steps C1 to C3 in the permanent magnet synchronous motor fault classification method.

[0031] The present invention provides a permanent magnet synchronous motor fault diagnosis, positioning and classification method and system. The diagnosis method and system extract the residual signal through the model predictive control (MPC) strategy, and perform coordinate transformation on the current residual to extract its DC component. The absolute difference of the DC component is used as a fault diagnosis indicator to determine whether a turn-to-turn short circuit fault or a high resistance connection fault has occurred. After determining that a fault has occurred, the positioning method and system further perform coordinate transformation based on the α- and β-axis residual currents, align with the B phase and the C phase, obtain fault positioning features of different phases, calculate the fault positioning index, and determine the phase where the fault occurs based on the fault positioning index. The classification method and system multiplies the α- and β-axis residual currents with the fault phase current and extracts the DC component to obtain classification features, calculates the fault classification index based on the classification features, classifies the fault according to the fault classification index, and determines whether the type of fault is a turn-to-turn short circuit fault or a high resistance connection fault.

[0032] Unlike traditional fault diagnosis methods based on current residuals, the present invention defines a fault diagnosis index by constructing the difference of the DC component of the current residual, and uses this index to classify faults. The present invention can determine whether two types of faults, namely inter-turn short circuit (ITF) and high resistance connection (HRC) faults, have occurred, and further determine the location and type of fault. It has high diagnostic, positioning and classification accuracy and robustness, and can adapt to complex working conditions such as speed, load changes and parameter mismatch. In addition, the fault diagnosis and classification process can be implemented online in real time and efficiently, with simple operation and strong real-time performance. The present invention does not require additional observers, hardware or control structure changes, can be easily embedded in existing motor control systems, and has broad industrial application potential. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a flow chart of a permanent magnet synchronous motor fault diagnosis method provided by an embodiment of the present invention;

[0034] Figure 2 This is a three-phase stator winding diagram of a permanent magnet motor with an inter-turn short circuit fault provided by an embodiment of the present invention;

[0035] Figure 3 This is a three-phase winding diagram of a permanent magnet motor with a high-resistance connection fault provided by an embodiment of the present invention;

[0036] Figure 4 This is a flow chart of a method for diagnosing, locating, and classifying permanent magnet synchronous motor faults provided by an embodiment of the present invention;

[0037] Figure 5 This is a diagram showing the experimental results of fault diagnosis using speed mutation, load mutation, and parameter mismatch tests provided by an embodiment of the present invention;

[0038] Figure 6 is a flow chart of a permanent magnet synchronous motor fault locating method provided by an embodiment of the present invention;

[0039] Figure 7 The embodiments of the present invention provide fault location indicators, fault location results, and fault current waveforms for different fault types and locations.

[0040] Figure 8 is a flow chart of a permanent magnet synchronous motor fault classification method provided by an embodiment of the present invention;

[0041] Figure 9 Schematic diagram of inter-turn short circuit fault and classification characteristics provided by an embodiment of the present invention;

[0042] Figure 10 This is a schematic diagram of high-resistance connection faults and classification characteristics provided by an embodiment of the present invention;

[0043] Figure 11 It is a diagnostic and classification performance diagram under different working conditions provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following describes the embodiments of the present invention in detail with reference to the accompanying drawings. The embodiments are provided for illustrative purposes only and are not to be construed as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of protection of the present invention. Many changes may be made to the present invention without departing from the spirit and scope of the present invention.

[0045] Example 1

[0046] The embodiment of the present invention provides a method for diagnosing permanent magnet synchronous motor faults, such as Figure 1 As shown in the flowchart, the steps include:

[0047] A1. Obtaining residual current: Under the model predictive control (MPC) framework, the residual currents of the α and β axes of the three-phase permanent magnet synchronous motor are calculated by the difference between the predicted currents of the α and β axes and the actual measured currents of the α and β axes.

[0048] A2. Calculate the fault diagnosis index: Multiply the α-axis and β-axis residual currents by themselves and extract their DC components through a low-pass filter (LPF). The absolute difference of the DC components is used as the fault diagnosis index.

[0049] A3. Perform fault diagnosis: In a fixed healthy state, 1.5 times the absolute difference between the DC components extracted after the α-axis and β-axis residual currents are multiplied by themselves is used as the fault diagnosis index threshold. The fault diagnosis index is compared with the fault diagnosis index threshold. If the fault diagnosis index is greater than the fault diagnosis index threshold, it is determined that a turn-to-turn short circuit fault or a high-resistance connection fault has occurred. Otherwise, it is determined to be normal.

[0050] Model predictive control (MPC) has been widely used in motor control due to its fast dynamic response, intuitive algorithm, and high flexibility. MPC strategies inherently possess error prediction and compensation capabilities, providing an effective means for extracting current residuals. The MPC-based residual signal not only reflects the motor's health but also captures the characteristic differences between turn-to-turn short circuits and high-resistance connection faults.

[0051] The three-phase stator winding of a permanent magnet motor with an inter-turn short circuit fault is as follows: Figure 2 As shown. In order to reflect the position of the fault phase, unlike the traditional modeling method, it is not assumed that the inter-turn short circuit fault occurs in a specific A, B or C phase, but it is assumed that the fault occurs in the fault phase, and the other two phases are the first normal phase and the second normal phase. The fault phase winding is divided into the normal winding L S1 and short-circuited winding L S2 , short-circuit coefficient μ=l S2 / (l S1 +l S2 ), where l S1 and l S2 They are normal winding L S1 and short-circuited winding L S1 The length, R f is the short-circuit resistance. The windings of the first normal phase and the second normal phase are represented by L h1 and L h2 ,i ITF_f is the fault current of the fault phase, i ph_f is the normal current of the fault phase, i h1 and i h2 are the currents of the first normal phase and the second normal phase respectively. ij represents the mutual inductance between windings i and j, i, j = S1, S2, h1 and h2 correspond to winding L respectively S1 、L S2 、L h1 and L h2 , and i≠j.

[0052] Under the condition of inter-turn short circuit fault, the voltage equations of the α and β axes of the permanent magnet motor are:

[0053]

[0054] Among them, i α 、i β They are α and β axis currents, R s is the stator resistance, θ e is the electrical angle, Ψ pm is the permanent magnet flux, ω e is the electrical angular velocity, L d 、L q They are the d-axis and q-axis inductances, μ represents, θ f is the electrical angle between the fault phase and motor phase A. If the fault occurs in phase A, θ f =0°. When the fault occurs in phase B or phase C, the f = -120° and θ f =-240°.

[0055] Due to the use of MPC control, the predicted current satisfies the voltage equation of a healthy motor. According to the voltage equation at ITF, the residual currents of the α and β axes at time k can be obtained by subtracting the predicted current from the measured current:

[0056]

[0057] Permanent magnet motor high resistance connection fault three-phase winding Figure 3 In order to reflect the location of the fault phase, unlike the traditional modeling method, it is not assumed that the high resistance connection fault occurs in a specific A, B or C phase. Instead, it is assumed that the fault occurs in the fault phase, and the other two phases are the first normal phase and the second normal phase. The currents of the fault phase, the first normal phase and the second normal phase are represented as i ph_f 、i h1 、i h2 ,i HRC_f is the fault current under HRC fault. According to the fault mechanism of high resistance connection, the fault phase is equivalent to a series fault resistance ΔR.

[0058] Under high resistance connection fault conditions, the d-axis and q-axis voltage equations of the permanent magnet motor are:

[0059]

[0060] i d 、i q They are d-axis and q-axis currents respectively.

[0061] Due to the use of MPC control, the predicted current satisfies the voltage equation of a healthy motor. According to the voltage equation during HRC, the residual current at time k can be obtained by subtracting the predicted current from the measured current:

[0062]

[0063] Among them, k-1 represents the moment before the kth moment, T s is the discrete period of the motor equation.

[0064] In order to maintain the consistency of the current residual form under HRC and the current residual under ITF, the current residual under d and q axes under HRC is subjected to inverse Park transformation to obtain:

[0065]

[0066] The corresponding DC components are obtained by multiplying the residual currents of α and β axes under HRC and ITF by themselves and filtering them:

[0067]

[0068] Among them, the parameters customized to simplify the formula form

[0069] By analyzing the above expressions, we can find that when the motor has the above two faults, the absolute difference between the α-axis and β-axis DC components is significantly larger. When the motor is in a healthy state, the absolute difference between the α-axis and β-axis DC components is relatively small. Therefore, this embodiment defines the fault diagnosis indicator as the absolute difference between the α-axis and β-axis DC components:

[0070]

[0071] They represent the DC components of the residual currents of the α and β axes of the permanent magnet synchronous motor at the actual time k.

[0072] By setting the fault diagnosis indicator threshold Thre dec , when FI dec >Thre dec When , it is determined to be a fault, that is, a turn-to-turn short circuit fault or a high resistance connection fault has occurred.

[0073] The specific flow chart of fault diagnosis is as follows: Figure 4 As shown, the DC components of the α and β axis residual currents of phases B and C are and Further obtain the fault diagnosis index FI of phase B and phase C dec_B , FI dec_C , similar to the fault diagnosis indicator threshold Thre dec Compare, if it is greater than the threshold Thredec , it is judged as a fault. The order of judgment is to first dec and the fault diagnosis indicator threshold Thre dec For comparison, if FI dec >Thre dec It is directly judged as a fault, otherwise FI dec_B and the fault diagnosis indicator threshold Thre dec For comparison, if FI dec_B >Thre dec It is directly judged as a fault, otherwise FI dec_C and the fault diagnosis indicator threshold Thre dec For comparison, if FI dec_C >Thre dec If the fault is detected, it is directly judged as a fault, otherwise it is judged as healthy.

[0074] The robustness of the proposed permanent magnet synchronous motor fault diagnosis method is verified through speed mutation, load mutation and parameter mismatch tests. The relevant experimental results are as follows: Figure 5 As shown, it shows that under different mutation conditions, and FI dec Changes from Figure 5 It can be seen that despite the sudden changes in speed, load, and parameters, the diagnostic indicators remain stable and no misdiagnosis occurs. In particular, in the case of sudden changes in speed, the diagnostic indicators remain unchanged, preventing the occurrence of misdiagnosis, demonstrating the strong robustness of the method.

[0075] In summary, the present embodiment provides a permanent magnet synchronous motor fault diagnosis method, which obtains the residual current information of the α and β axes of the motor through a model predictive control (MPC) framework, performs coordinate transformation on the current residual, extracts its DC component, and further defines a fault diagnosis index by constructing the difference of the DC component of the current residual. The index is used to perform fault diagnosis and determine whether a fault occurs. The method has a faster diagnostic speed and higher diagnostic accuracy and robustness.

[0076] Example 2

[0077] This embodiment provides a permanent magnet synchronous motor fault diagnosis system, comprising a residual current acquisition module, a fault diagnosis index calculation module, and a fault diagnosis module, each of which is configured to execute steps A1 to A3 of the permanent magnet synchronous motor fault diagnosis method described in Example 1. The specific functions of the residual current acquisition module, the fault diagnosis index calculation module, and the fault diagnosis module have been described in detail in Example 1 and will not be repeated in this embodiment. This embodiment emphasizes the existence of an electronic system capable of diagnosing permanent magnet synchronous motor faults using the permanent magnet synchronous motor fault diagnosis method described in Example 1.

[0078] Example 3

[0079] Based on the determination of the occurrence of a fault in Example 1, it is necessary to further locate the fault. This embodiment provides a method for locating a fault of a permanent magnet synchronous motor, such as Figure 6 As shown in the flowchart, the steps include:

[0080] B1. Perform coordinate transformation on the α-axis and β-axis residual currents to rotate them to align with phases B and C, and obtain the residual currents in the coordinate system aligned with phases B and C.

[0081] B2. Squaring the rotated residual current with itself and extracting the DC component through low-pass filtering, and using it together with the DC components of the α-axis and β-axis residual currents as fault location features;

[0082] B3. Calculate the fault location index based on the fault location characteristics;

[0083] B4. Locate the fault according to the fault location indicators and determine the location where the fault occurred.

[0084] The characteristics of different fault locations under ITF and HRC are shown in Table 1.

[0085] Table 1 Fault characteristics of different fault types and fault locations

[0086]

[0087] To determine the fault phase, the residual currents of the α and β axes are transformed to align with different phases. The residual currents in the α and β coordinate systems are expressed as By performing a rotation transformation using the Euler formula, we can obtain the residual current at time k in the coordinate system aligned with phases B and C:

[0088]

[0089] The rotated residual current is squared by itself and the DC component is extracted through low-pass filtering to obtain the corresponding B-phase and C-phase fault location features:

[0090]

[0091] The residual currents of the α and β axes at time k are multiplied by themselves and the DC component is extracted through low-pass filtering to obtain:

[0092]

[0093] This embodiment defines the fault location indicators as:

[0094]

[0095] Then combine the fault location indicators of different coordinate systems, such as Figure 4 As shown in the flowchart, the fault location decision logic is as follows:

[0096] If FI loc_A >0 and FI loc_B >0 or FI loc_A <0 and FI loc_B <0, the fault is determined to occur in phase C;

[0097] If FI loc_A >0 and FI loc_C <0, it is determined that the fault occurs in phase A;

[0098] If FI loc_A <0 and FI loc_B >0, it is determined that the fault occurs in phase B.

[0099] This logic is based on the residual characteristics of different fault locations and can effectively distinguish the phases in which the fault occurs.

[0100] Fault location indicators, fault location results and fault current waveforms under different fault types and locations are as follows: Figure 7 As shown, Figure 7 (a) corresponds to the location result of phase A (ITF fault), Figure 7 (b) corresponds to the positioning result of phase C (HRC fault). By simulating the ITF fault of phase A and the HRC fault of phase C, Figure 7 Experimental results show that the fault localization index changes suddenly after a fault occurs, and by comparing the fault localization index of each phase, the fault phase can be accurately determined. The delay of the localization process is 63.1ms (ITF) and 329.2ms (HRC), respectively, indicating that the method has good real-time performance and localization accuracy.

[0101] In summary, the present embodiment provides a permanent magnet synchronous motor fault location method, which performs coordinate transformation on the α- and β-axis residual currents to obtain the residual current in a coordinate system aligned with the B-phase and C-phase. The rotated residual current is then multiplied by itself and the DC component is extracted through low-pass filtering. The DC component and the DC component of the α- and β-axis residual currents are used together as fault location features. The fault location index is calculated based on the fault location feature, and the fault is located based on the fault location index, so that the location of the inter-turn short circuit (ITF) fault or the high-resistance connection (HRC) fault can be accurately determined. Experiments have verified the effectiveness and excellence of the present method.

[0102] Example 4

[0103] This embodiment provides a permanent magnet synchronous motor fault location system, comprising a coordinate transformation module, a fault location feature calculation module, a fault location index calculation module, and a fault location module, each of which is configured to execute steps B1 through B4 of the permanent magnet synchronous motor fault diagnosis method described in Example 3. The specific functions of the coordinate transformation module, the fault location feature calculation module, the fault location index calculation module, and the fault location module have been described in detail in Example 3 and will not be repeated in this embodiment. This embodiment emphasizes the existence of an electronic system capable of locating permanent magnet synchronous motor faults using the permanent magnet synchronous motor fault location method described in Example 3.

[0104] Example 5

[0105] Based on the fault location determined in Example 3, it is necessary to classify the fault type to determine the type of fault that occurred. This embodiment provides a permanent magnet synchronous motor fault classification method, such as Figure 8 As shown in the flowchart, the method further comprises the steps of:

[0106] C1. Multiply the α-axis and β-axis residual currents with the fault phase current and extract their DC components through a low-pass filter to obtain classification features.

[0107] C2. Calculate fault classification index based on classification features;

[0108] C3. Classify the fault according to the fault classification index to determine whether the fault type is a turn-to-turn short circuit fault or a high-resistance connection fault.

[0109] Fault classification is based on the phase characteristics of the α-axis and β-axis residual currents. The classification characteristics are obtained by multiplying the α-axis and β-axis residual currents with the fault phase current and extracting their DC components through a low-pass filter:

[0110]

[0111] The fault classification index is defined as the difference between the classification features:

[0112]

[0113] Where θ0 is the angle between the fault current and the fault phase current at ITF. Under HRC conditions, FI cla is a positive value; under ITF conditions, FI cla is a negative value.

[0114] The fault classification process is as follows Figure 4 As shown, set the fault classification threshold Thre cla =0, when FI cla >Thre cla It is judged as HRC when it is detected, otherwise it is judged as ITF.

[0115] Related fault classification characteristics are shown in the figure Figure 9 and Figure 10 , Figure 9 is the inter-turn short circuit fault and classification characteristics, Figure 10 High resistance connection fault and classification characteristics. Figure 9 (a) corresponds to the A-phase ITF fault, Figure 9 (b) corresponds to the B-phase ITF fault, Figure 9 (c) corresponds to the C phase ITF fault. Figure 10 (a) corresponds to the A phase HRC fault, Figure 10 (b) corresponds to the B phase HRC fault, Figure 10 (c) corresponds to the C phase HRC fault.

[0116] To further verify the effectiveness of the method, experiments were conducted in the range of 300 to 600 rpm and 30% to 50% of the rated load. Figure 11 , Figure 11 (a) is the performance of diagnostic indicators under various working conditions, Figure 11 (b) is the performance of classification indicators. Figure 11 Results show that in a healthy state, diagnostic indicators are largely unaffected by changes in operating conditions, allowing for fixed thresholds. In the ITF case, diagnostic indicators increase with increasing speed and load. In the HRC case, increasing speed reduces the effect of resistance on the total impedance, but increased phase current enhances the HRC fault characteristics. Experimental results demonstrate that this approach works effectively under a wide range of operating conditions.

[0117] In summary, this embodiment provides a permanent magnet synchronous motor fault classification method. This method multiplies the α- and β-axis residual currents with the fault phase current and extracts its DC component through a low-pass filter to obtain classification features. A fault classification index is then calculated based on the classification features. Fault classification is then performed based on the fault classification index to determine whether the fault is a turn-to-turn short circuit or a high-resistance connection fault. Combined with the permanent magnet synchronous motor fault location method provided in Example 3, this method can accurately determine the fault type and fault phase. Experimental results demonstrate that this method not only rapidly identifies faults but also accurately locates the fault phase, with good real-time performance for both ITF and HRC fault types. Further operating condition verification demonstrates that this method maintains stable operation under varying speed and load conditions and exhibits strong robustness, effectively addressing complex situations such as sudden speed changes, load variations, and parameter mismatches. The proposed method demonstrates high accuracy, low latency, and strong robustness under a variety of operating conditions, possessing broad practical application value.

[0118] Example 6

[0119] This embodiment provides a permanent magnet synchronous motor fault classification system, comprising a fault classification feature calculation module, a fault classification index calculation module, and a fault classification module, each of which is configured to execute steps C1 through C3 of the permanent magnet synchronous motor fault classification method described in Example 5. The specific functions of the fault classification feature calculation module, the fault classification index calculation module, and the fault classification module have been described in detail in Example 5 and will not be repeated in this embodiment. This embodiment emphasizes the existence of an electronic system capable of classifying permanent magnet synchronous motor faults using the permanent magnet synchronous motor fault classification method described in Example 5.

[0120] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be considered as equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. A method for locating a fault of a permanent magnet synchronous motor, characterized in that: Including steps: A1. Calculate the residual current of the α and β axes by the difference between the predicted current of the α and β axes and the actual measured current of the α and β axes of the three-phase permanent magnet synchronous motor; A2. Squaring the residual currents of the α and β axes and extracting their DC components, and using the absolute difference of the DC components as a fault diagnosis indicator; A3. 1.5 times the absolute difference between the DC components of the α-axis and β-axis residual currents, extracted after multiplying them by themselves, is used as the fault diagnosis indicator threshold. The fault diagnosis indicator is compared with the fault diagnosis indicator threshold. If the fault diagnosis indicator is greater than the fault diagnosis indicator threshold, it is determined that a turn-to-turn short circuit fault or a high-resistance connection fault has occurred. Otherwise, it is determined to be normal. After it is determined in step A3 that an inter-turn short circuit fault or a high resistance connection fault has occurred, the process includes the following steps: B1. Perform coordinate transformation on the residual currents of the α and β axes to rotate them to align with the B and C phases, and obtain the residual currents in the coordinate system aligned with the B and C phases; B2. Squaring the rotated residual current with itself and extracting the DC component through low-pass filtering, and using it together with the DC components of the α-axis and β-axis residual currents as fault location features; B3. Calculate the fault location index based on the fault location characteristics; B4. Locate the fault according to the fault location indicators and determine the location where the fault occurred.

2. The permanent magnet synchronous motor fault location method according to claim 1, characterized in that: In step B3, the fault location indicators include: ； in, It represents the residual current of α and β axes at time k and extracts the DC component through low-pass filtering. 、 difference, The residual current of phase B after rotation at time k is squared and the DC component is extracted by low-pass filtering. 、 difference, The C-phase residual current after rotation at time k is squared and the DC component is extracted by low-pass filtering. 、 difference.

3. The permanent magnet synchronous motor fault location method according to claim 2, characterized in that: The step B4 is specifically as follows: if >0 and >0 or <0 and <0, the fault is determined to occur in phase C; if >0 and <0, it is determined that the fault occurs in phase A; if <0 and >0, it is determined that the fault occurs in phase B.

4. Permanent magnet synchronous motor fault location system, characterized by: It includes a coordinate transformation module, a fault location feature calculation module, a fault location index calculation module, and a fault location module, which are respectively used to execute steps B1 to B4 in the permanent magnet synchronous motor fault location method according to any one of claims 1 to 3.

5. A permanent magnet synchronous motor fault classification method, based on the permanent magnet synchronous motor fault location method according to claim 1, characterized in that: After determining in step A3 that an inter-turn short circuit fault or a high resistance connection fault has occurred, the classification method includes the following steps: C1. Multiply the α-axis and β-axis residual currents with the fault phase current and extract their DC components through a low-pass filter to obtain classification features. C2. Calculate fault classification index based on classification features; C3. Classify the fault according to the fault classification index to determine whether the fault type is a turn-to-turn short circuit fault or a high-resistance connection fault.

6. The permanent magnet synchronous motor fault classification method according to claim 5, characterized in that: In step C2, the fault classification index is the α and β axis classification features FI cla_α , FI cla_β The difference between FI cla =FI cla_α -FI cla_β .

7. The permanent magnet synchronous motor fault classification method according to claim 6, characterized in that: The step C3 is specifically as follows: setting the fault classification threshold =0, when > If the fault is positive, it is determined as a high-resistance connection fault; otherwise, it is determined as an inter-turn short-circuit fault.

8. Permanent magnet synchronous motor fault classification system, characterized by: It includes a fault classification feature calculation module, a fault classification index calculation module, and a fault classification module, which are respectively used to execute steps C1 to C3 in the permanent magnet synchronous motor fault classification method according to any one of claims 5 to 7.

Images