Permanent magnet synchronous motor turn-to-turn short circuit and inverter open circuit fault detection and distinguishing method

Abstract

The invention provides a permanent magnet synchronous motor turn-to-turn short circuit and inverter open circuit fault detection and distinguishing method, and belongs to the technical field of motor driving system fault detection. The technical problems that in the prior art, fault detection is poor in real-time performance, and fault types are difficult to distinguish are solved. According to the technical scheme, three-phase stator current signals of the motor are collected in real time, data of N recent sampling points are dynamically stored, and a current data set updated in real time is formed. And in each diagnosis period, calculating a Pearson correlation coefficient between every two three-phase currents based on the data set, generating a comprehensive diagnosis variable P, monitoring the change of a P value in real time, triggering a fault alarm when the P exceeds a preset threshold value, and distinguishing turn-to-turn short circuit and inverter open circuit faults according to the characteristic mode of the P value. The method has the advantages that the operation reliability of the permanent magnet synchronous motor system is improved, and the technology aims at achieving real-time monitoring of turn-to-turn short circuit and inverter open circuit faults of the permanent magnet synchronous motor system.

Description

Technical Field

[0001] The present invention relates to the technical field of motor drive system fault detection, and in particular to a method for detecting and distinguishing between inter-turn short circuit faults of a permanent magnet synchronous motor and open circuit faults of an inverter. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in electric vehicles, industrial drives, and renewable energy due to their high efficiency, high power density, and excellent dynamic performance. However, under complex operating conditions, motor drive systems can experience a variety of faults, two common of which are stator winding interturn short circuits (ISCs) and inverter open-circuit faults (OCFs). Interturn short circuits can cause local overheating, increased electromagnetic torque fluctuations, and even more seriously, winding burnout; while inverter open-circuit faults can cause motor torque pulsation, speed fluctuations, and even system loss of control. Therefore, quickly and accurately detecting and distinguishing these two types of faults is crucial to improving system reliability and safety.

[0003] Existing fault detection methods are primarily based on signal analysis, model observation, or artificial intelligence techniques, such as current harmonic analysis, voltage residual detection, and neural network classification. However, these methods often suffer from computational complexity, high parameter sensitivity, or difficulty distinguishing similar fault characteristics. In particular, turn-to-turn short-circuit and converter open-circuit faults can exhibit similar trends in current waveforms, making it difficult for traditional methods to effectively distinguish them. In recent years, methods based on statistical correlation, such as the Pearson correlation coefficient, have shown potential in fault diagnosis due to their computational simplicity and robustness to noise. However, their application to the detection and differentiation of complex faults in PMSM systems remains under investigation. Therefore, there is an urgent need to develop a fault detection and differentiation method based on Pearson correlation analysis to achieve efficient and accurate identification of PMSM turn-to-turn short-circuit and converter open-circuit faults, providing reliable technical support for the health management of motor drive systems. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for detecting and distinguishing inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor, thereby improving the operating reliability of the permanent magnet synchronous motor system. This technology aims to achieve real-time monitoring of inter-turn short circuit and inverter open circuit faults of the permanent magnet synchronous motor system.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts a technical solution as follows: a method for detecting and distinguishing between inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor, comprising the following steps:

[0006] S1. In each control cycle, the three-phase stator current signal of the permanent magnet synchronous motor system is synchronously collected at a fixed sampling frequency, and the data of the most recent N sampling points are stored to form a dynamically updated current data set;

[0007] S2. In each diagnostic cycle, based on the latest stored data set, calculate the Pearson correlation coefficient between the three-phase current signals and generate a comprehensive diagnostic variable P according to the preset rules;

[0008] S3, real-time detection of diagnostic variable P. When P exceeds the normal threshold range, a fault alarm is triggered and the permanent magnet synchronous motor stator winding inter-turn short circuit fault and the inverter IGBT open circuit fault are distinguished based on P.

[0009] Furthermore, the step S1 includes the following steps:

[0010] Step 1.1) Set the control period to T c , with a fixed sampling frequency f in each control cycle s Synchronously collect the three-phase stator current signal i of the permanent magnet synchronous motor system a 、i b 、i c , where the sampling frequency f s To satisfy the Nyquist sampling theorem and take into account both accuracy and real-time performance, it is set to 10 times the fundamental frequency f1, that is, f s =10f1;

[0011] Step 1.2) Set the data window length N = k n ·f s T1, where k n represents an integer greater than or equal to 1, and T1 represents the current fundamental wave period.

[0012] Step 1.3) The collected three-phase current signals are stored in a dynamically updated current data set I in chronological order, where I = [i a ,i b ,i c ], and i a =[i a (1),i a (2),…,i a (N)],i b =[i b (1),i b (2),…,i b (N)],i c =[i c (1),i c (2),…,i c (N)], which are the current data of phase A, phase B and phase C at the latest N sampling points;

[0013] Step 1.4) A first-in-first-out data update strategy is used. When each new sampling point arrives, the oldest data point is removed from the data set and the latest sampling value is stored in it, ensuring that I always contains the latest three-phase current data of the N sampling points;

[0014] Furthermore, step S2 includes the following steps:

[0015] Step 2.1) Calculate the absolute value of the Pearson correlation coefficient between the phase currents based on the data set I ab 、p bc 、p ca (between phase A and phase B, between phase B and phase C, between phase C and phase A), which is used to measure the correlation strength of the two-phase current signals. The calculation formula is as follows:

[0016]

[0017] Where x,y∈{a,b,c}, and x≠y, that is, calculate p ab 、p bc 、p ca ; cov represents covariance; σ represents standard deviation; E represents mean.

[0018] Step 2.2) Based on p ab 、p bc 、p ca Calculate the diagnostic variable P using the following formula:

[0019]

[0020] Furthermore, step S3 includes the following steps:

[0021] Step 3.1) By comparing the diagnostic variable P with the preset fault threshold P th Comparison of the system state is used to determine the system state. When P≥P th When P <P th When the system is in a healthy state, it is judged that the system is in a healthy state and the operation monitoring is continued. th The setting is determined according to the fault tolerance characteristics of the system, and the specific quantitative standard is: P th =0.5×P| 5％_turn_fault , that is, 50% of the P value when 5% of the system has inter-turn faults;

[0022] Step 3.2) After the fault is confirmed, compare the diagnostic variable P with the fault discrimination threshold P div Perform a secondary comparison, if P>P div , it is determined to be an inverter open circuit fault; if P th ≤P≤Pdiv , it is determined to be a permanent magnet synchronous motor turn-to-turn short circuit fault. div is significantly greater than P th The positive threshold value of .

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1. The present invention discloses a method for detecting and distinguishing inter-turn short-circuit faults in permanent magnet synchronous motors and open-circuit faults in inverters. This method has high computational efficiency and only requires a simple Pearson correlation coefficient analysis of the phase current signals, avoiding the computational burden of complex algorithms and is suitable for real-time monitoring of embedded systems.

[0025] 2. Its robustness effectively suppresses noise interference, maintains stable performance under different operating conditions, and significantly reduces false alarm rates. Furthermore, the method is highly accurate. By capturing subtle differences in current correlation characteristics, it can accurately distinguish between turn-to-turn short-circuit and converter open-circuit faults, reliably identifying even early-stage turn-to-turn short-circuit faults. These characteristics make this method highly practical in engineering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0027] Figure 1 This is a flow chart of the fault detection and differentiation strategy in the present invention.

[0028] Figure 2 This is a curve diagram of the diagnostic variable P when a 10% inter-turn short circuit fault occurs in the permanent magnet synchronous motor of the present invention.

[0029] Figure 3 This is a curve diagram of the diagnostic variable P when a single-tube open-circuit fault occurs in the inverter of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0031] Example 1

[0032] See also Figure 1-Figure 3 This embodiment provides a technical solution based on a method for detecting and distinguishing between inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor: Figure 1The strategic process of the fault detection and differentiation method is presented. In each control cycle, the three-phase stator current signals of the motor are synchronously collected at a fixed sampling frequency, and the data of the most recent N sampling points are dynamically stored to form a real-time updated current data set I. In each diagnostic cycle, the Pearson correlation coefficient between the three-phase currents is calculated based on I, and a comprehensive diagnostic variable P is generated. By monitoring the changes in the P value in real time, when P exceeds the preset threshold P th The fault alarm is triggered when the fault occurs, and the inter-turn short circuit and inverter IGBT open circuit fault are distinguished based on the characteristic pattern of the P value.

[0033] The present invention provides a method for detecting and distinguishing between inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor, the method comprising the following steps:

[0034] S1, in each control cycle, synchronously collects the three-phase stator current signals of the permanent magnet synchronous motor system at a fixed sampling frequency and stores the data of the most recent N sampling points to form a dynamically updated current data set;

[0035] S2, in each diagnostic cycle, calculates the Pearson correlation coefficient between the three-phase current signals based on the latest stored data set, and generates a comprehensive diagnostic variable P according to the preset rules;

[0036] S3, real-time detection of diagnostic variable P. When P exceeds the normal threshold range, a fault alarm is triggered and the permanent magnet synchronous motor stator winding inter-turn short circuit fault and the inverter IGBT open circuit fault are distinguished based on P.

[0037] Furthermore, the specific method of step S1 is as follows:

[0038] Step 1.1) Set the control period to T c , with a fixed sampling frequency f in each control cycle s Synchronously collect the three-phase stator current signal i of the permanent magnet synchronous motor system a 、i b 、i c , where the sampling frequency f s To satisfy the Nyquist sampling theorem and take into account both accuracy and real-time performance, it is set to 10 times the fundamental frequency f1, that is, f s =10f1;

[0039] Step 1.2) Set the data window length N = k n ·f s T1, where k n represents an integer greater than or equal to 1, and T1 represents the current fundamental wave period.

[0040] Step 1.3) The collected three-phase current signals are stored in a dynamically updated current data set I in chronological order, where I = [i a,i b ,i c ], and i a =[i a (1),i a (2),…,i a (N)],i b =[i b (1),i b (2),…,i b (N)],i c =[i c (1),i c (2),…,i c (N)], which are the current data of phase A, phase B and phase C at the latest N sampling points;

[0041] Step 1.4) A first-in-first-out data update strategy is used. When each new sampling point arrives, the oldest data point is removed from the data set and the latest sampling value is stored in it, ensuring that I always contains the latest three-phase current data of the N sampling points;

[0042] Furthermore, the specific method of step S2 is as follows:

[0043] Step 2.1) Calculate the absolute value of the Pearson correlation coefficient between the phase currents based on the data set I ab 、p bc 、p ca (between phase A and phase B, between phase B and phase C, between phase C and phase A), which is used to measure the correlation strength of the two-phase current signals. The calculation formula is as follows:

[0044]

[0045] Where x,y∈{a,b,c}, and x≠y, that is, calculate p ab 、p bc 、p ca ; cov represents covariance; σ represents standard deviation; E represents mean.

[0046] Step 2.2) Based on p ab 、p bc 、p ca Calculate the diagnostic variable P using the following formula:

[0047]

[0048] Table 1 shows the range of diagnostic variable P under different fault conditions, including three fault conditions: 5%, 10%, and 20% inter-turn short-circuit faults in permanent magnet synchronous motors, as well as four typical inverter open-circuit faults. The analysis results show that: 1) the diagnostic variable P shows a significant increase after the fault occurs; 2) in inter-turn short-circuit faults, the P value is positively correlated with the fault severity, increasing with the severity of the short-circuit; 3) while the P value under open-circuit faults fluctuates within a certain range, its value is significantly higher than that under inter-turn short-circuit faults, generally by an order of magnitude.

[0049] Table 1 Variation range of diagnostic variable P

[0050]

[0051] Furthermore, the specific method of step S3 is as follows:

[0052] Step 3.1) By comparing the diagnostic variable P with the preset fault threshold P th Comparison of the system state is used to determine the system state. When P≥P th When P <P th When the system is in a healthy state, it is judged that the system is in a healthy state and the operation monitoring is continued. th The setting of is determined according to the fault tolerance characteristics of the system. In this embodiment, P th =0.015.

[0053] Step 3.2) After the fault is confirmed, compare the diagnostic variable P with the fault discrimination threshold P div Perform a secondary comparison, if P>P div , it is determined to be an inverter open circuit fault; if P th ≤P≤P div , it is determined to be a permanent magnet synchronous motor turn-to-turn short circuit fault. div is significantly greater than P th The positive threshold value of .

[0054] Figure 2 and Figure 3 The curves of the diagnostic variable P before and after a 10% interturn short-circuit fault in a permanent magnet synchronous motor and a single-switch open-circuit fault in the inverter are shown. The graph clearly shows that during normal system operation, the P value remains close to zero; however, when a fault occurs, the P value increases significantly. Furthermore, the increase in P value due to an open-circuit fault is much greater than that due to an interturn short-circuit fault, typically by more than an order of magnitude. Therefore, the diagnostic variable P can not only effectively distinguish between normal and faulty system states, but also between interturn short-circuit faults in permanent magnet synchronous motors and open-circuit faults in inverters.

[0055] Example 2

[0056] On the same permanent magnet synchronous motor test platform, the fault detection performance of the proposed Pearson correlation coefficient method was compared with that of a traditional harmonic analysis method (FFT-based fifth harmonic amplitude detection). Test conditions included a healthy state (no faults), a 10% turn-to-turn short-circuit fault, and an inverter single-transistor open-circuit fault.

[0057] Table 2 Comparison of traditional harmonic analysis methods

[0058]

[0059] The experimental results are shown in Table 2, and the following experimental conclusions can be drawn:

[0060] 1. Stronger anti-interference ability: When the load changes suddenly, the amplitude fluctuation range of the fifth harmonic of the traditional harmonic analysis method reaches ±40%, which is easy to cause false alarms; while the P value output by the method of the present invention fluctuates within a range of <±5%, which significantly improves the detection stability.

[0061] 2. Fault differentiation is more direct: Traditional methods need to rely on additional logic to determine the fault type, while the present invention can directly distinguish turn-to-turn short-circuit faults (P≤P div ) and inverter open circuit fault (P>P div ), simplifying the troubleshooting process.

[0062] This embodiment verifies that the method of the present invention is superior to the traditional harmonic analysis method in terms of anti-interference and fault differentiation capabilities, and is particularly suitable for motor fault detection under high dynamic load conditions.

[0063] Example 3

[0064] The following two typical faults were tested on the same permanent magnet synchronous motor experimental platform: a 10% inter-turn short circuit fault and a single-tube open circuit fault. The test conditions covered a variety of operating conditions to verify the robustness of the proposed method:

[0065] Load variation: 30%, 70%, 100% rated load

[0066] Speed ​​changes: 500rpm, 1500rpm, 3000rpm

[0067] Noise interference: Add 15dB Gaussian white noise to the current signal (simulating the actual industrial environment)

[0068] Number of experiments: Repeat 100 times for each working condition

[0069] Table 3 Comparison of fault detection success rates between the method of the present invention and the traditional residual current method

[0070]

[0071] The experimental results are shown in Table 3, and the following experimental conclusions can be drawn:

[0072] 1. High reliability: The method of the present invention maintains a detection success rate of >97% under harsh working conditions (high load, high speed, noise interference), which is significantly better than the traditional residual current method (<85%).

[0073] 2. Optimizability: Remaining misjudgments can be further eliminated through hardware synchronization optimization or adaptive threshold strategies, which has engineering optimization potential.

[0074] This embodiment verifies that the method of the present invention still has high-precision and strong robust fault detection capabilities in complex industrial environments, and is suitable for real-time monitoring and protection of actual motor systems.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting and distinguishing between inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor, characterized in that: The following steps are involved: S1. In each control cycle, the three-phase stator current signal of the permanent magnet synchronous motor system is synchronously collected at a fixed sampling frequency, and the data of the most recent N sampling points are stored to form a dynamically updated current data set; S2. In each diagnostic cycle, based on the latest stored data set, calculate the Pearson correlation coefficient between the three-phase current signals and generate a comprehensive diagnostic variable P according to the preset rules; S3, real-time detection of diagnostic variable P. When P exceeds the normal threshold range, a fault alarm is triggered and the permanent magnet synchronous motor stator winding inter-turn short circuit fault and the inverter IGBT open circuit fault are distinguished based on P.

2. A method for detecting and distinguishing between inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor according to claim 1, characterized in that: The step S1 includes the following steps: Step 1.1) Set the control period to T c , with a fixed sampling frequency f in each control cycle s Synchronously collect the three-phase stator current signal i of the permanent magnet synchronous motor system a 、i b 、i c , where the sampling frequency f s To satisfy the Nyquist sampling theorem and take into account both accuracy and real-time performance, it is set to 10 times the fundamental frequency f1, that is, f s =10f1; Step 1.2) Set the data window length N = k n ·f s T1, where k n represents an integer greater than or equal to 1, T1 represents the current fundamental wave period; Step 1.3) The collected three-phase current signals are stored in a dynamically updated current data set I in chronological order, where I = [i a ,i b ,i c ], and i a =[i a (1),i a (2),…,i a (N)],i b =[i b (1),i b (2),…,i b (N)],i c =[i c (1),i c (2),…,i c (N)], which are the current data of phase A, phase B and phase C at the latest N sampling points; Step 1.4) adopts a first-in-first-out data update strategy. When each new sampling point arrives, the oldest data point is removed from the data set and the latest sampling value is stored in it, ensuring that I always contains the latest three-phase current data of the N sampling points.

3. A method for detecting and distinguishing between inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor according to claim 2, characterized in that: The step S2 comprises the following steps: Step 2.1) Calculate the absolute value of the Pearson correlation coefficient between the phase currents based on the data set I ab 、p bc 、p ca , which is used to measure the correlation strength of the two-phase current signals. Its calculation formula is as follows: Where x,y∈{a,b,c}, and x≠y, that is, calculate p ab 、p bc 、p ca ; cov represents covariance; σ represents standard deviation; E represents the mean; Step 2.2) Based on p ab 、p bc 、p ca Calculate the diagnostic variable P using the following formula:

4. A method for detecting and distinguishing between inter-turn short circuit and inverter open circuit faults of a permanent magnet synchronous motor according to claim 3, characterized in that: The step S3 comprises the following steps: Step 3.1) By comparing the diagnostic variable P with the preset fault threshold P th Comparison of the system state is used to determine the system state. When P≥P th When P <P th When the system is judged to be in a healthy state, the operation monitoring is continued, wherein the fault threshold P th The setting is determined according to the system fault tolerance characteristics; Step 3.2) After the fault is confirmed, compare the diagnostic variable P with the fault discrimination threshold P div Perform a secondary comparison, if P>P div , it is determined to be an inverter open circuit fault; if P th ≤P≤P div , it is determined to be a permanent magnet synchronous motor turn-to-turn short circuit fault, wherein the fault distinction threshold P div is significantly greater than P th The positive threshold value of .

Images