Method for improving reliability of early turn - to - turn short - circuit fault diagnosis of permanent magnet synchronous motor

Abstract

The method for improving the reliability of early turn-to-turn short circuit fault diagnosis of a permanent magnet synchronous motor disclosed by the present invention is specifically implemented according to the following steps: Step 1: Establish a mathematical model of the turn-to-turn short circuit fault of the permanent magnet synchronous motor; Step 2: Design a voltage disturbance observer for compensating the power supply imbalance caused by the compensated inverter according to the mathematical model of the turn-to-turn short circuit fault of the permanent magnet synchronous motor; Step 3: After compensation, extract the high-frequency current response under the injection of high-frequency signals of the permanent magnet synchronous motor through a band-pass filter, and then perform coordinate transformation and low-pass filter to extract the fault characteristics. This method solves the problem that the fault characteristics in the early stage of the turn-to-turn short circuit fault are not obvious and the fault cannot be accurately detected.

Description

Technical Field

[0001] The present invention belongs to the technical field of permanent magnet synchronous motor fault diagnosis, and specifically relates to a method for improving the reliability of early turn - to - turn short - circuit fault diagnosis of permanent magnet synchronous motors. Background Art

[0002] Permanent Magnet Synchronous Motors (PMSMs) have many advantages such as high power density, high efficiency, and high torque density, and have been widely used in industrial production, daily life and other occasions. Since PMSMs are prone to various faults during operation due to factors such as narrow working environment space, high temperature, high humidity, poor heat dissipation conditions, as well as mechanical and electrical factors. Among them, turn - to - turn short - circuit faults are the most common type of faults, which are highly destructive and likely to cause other faults. After a turn - to - turn short - circuit fault occurs, if it cannot be detected and corresponding measures are not taken in the initial stage of the fault, more serious consequences will eventually occur. To improve the safety and reliability of PMSMs, early turn - to - turn short - circuit fault diagnosis of PMSMs is very important. When the turn - to - turn short - circuit fault is in the early stage, its fault characteristics are not obvious and are easily submerged in other harmonic signals and noises, making it difficult to extract, which poses a great challenge to diagnosis.

[0003] Injecting high - frequency voltage signals can amplify fault characteristics. When a turn - to - turn short - circuit fault occurs, a short - circuit current will appear in the short - circuited winding, and the short - circuit current will form a pulsating magnetomotive force in the air gap, which affects the high - frequency current generated by the high - frequency voltage. The fault characteristics used in this patent are highly sensitive to turn - to - turn short - circuit faults and less sensitive to the flux level and load level. Therefore, using them as fault characteristics will greatly improve the reliability of early diagnosis of turn - to - turn short - circuit faults. Since non - ideal behaviors such as power supply imbalance caused by inverter dead - time will affect the high - frequency current response and lead to a decrease in the diagnostic accuracy, compensating for the dead - time can improve the diagnostic reliability. Current compensation methods include disturbance observer compensation, current feedback compensation, voltage feedback compensation, etc. Among them, the dead - time compensation based on the disturbance observer has simple calculations and does not require accurate knowledge of the DC - side voltage and dead - time. Online compensation can be achieved through the observer, which does not affect the system stability and does not require additional hardware, and can be implemented only through software. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for improving the reliability of early turn - to - turn short - circuit fault diagnosis of permanent magnet synchronous motors, which solves the problem that the fault characteristics in the early stage of turn - to - turn short - circuit faults are not obvious and the faults cannot be accurately detected.

[0005] The technical solution adopted by the present invention is that the method for improving the reliability of early turn - to - turn short - circuit fault diagnosis of permanent magnet synchronous motors is specifically implemented according to the following steps:

[0006] Step 1: Establish a mathematical model for the inter-turn short circuit fault of a permanent magnet synchronous motor;

[0007] Step 2: Design a voltage disturbance observer that compensates for the power supply imbalance caused by the compensation inverter according to the mathematical model of the inter-turn short circuit fault of the permanent magnet synchronous motor;

[0008] Step 3: After compensation, extract the high-frequency current response under the injection of high-frequency signals of the permanent magnet synchronous motor through a band-pass filter, and then perform coordinate transformation and low-pass filter to extract the fault characteristics.

[0009] The features of the present invention also lie in that,

[0010] Step 1 is specifically as follows:

[0011] When an inter-turn short circuit fault occurs in phase A of the motor, from Figure 1 it can be seen that a short circuit loop will be added to the winding of phase A of the motor. At this time, the resistance r f divides phase A into a healthy part and a faulty part, and the short-circuit turn ratio μ is defined as:

[0012]

[0013] where N f is the number of short-circuited turns of a certain phase of the stator winding, and N is the total number of turns of a certain phase of the stator winding;

[0014] Establish a motor mathematical model according to the voltage equation of the faulty permanent magnet synchronous motor:

[0015]

[0016] where V abcf = [V ah V b V c V af T is the phase voltage matrix of the stator winding, where V ah , V b , V c , V af are the voltages of the healthy part of phase A, the voltages of phases B and C, and the voltage of the faulty part of phase A respectively; i abcf = [i a i b i c i f T is the current matrix, i a , i b , r c , i f are the three-phase stator currents and the short-circuit current respectively; e m = [e ah e​​b e c e af T is the back electromotive force matrix of the three-phase stator winding and the short-circuit winding, where e ah and e b and e c and e af are the back electromotive force of the healthy part of phase A, the back electromotive forces of phases B and C, and the back electromotive force of the faulty part of phase A, respectively;

[0017] Equation (2) is transformed through coordinate transformation to obtain the voltage equation in the d-q axis system:

[0018]

[0019] where, u d and u q are the d-axis and q-axis voltages, i d and i q are the d-axis and q-axis currents, L d and L q are the d-axis and q-axis inductances, ω is the motor speed, θ r is the rotor position angle, and λ is the amplitude of the stator magnetic flux.

[0020] In step 1:

[0021]

[0022] Equation (3) is the resistance matrix, where R s is the stator resistance, r f is the resistance of the faulty winding A f ;

[0023]

[0024] Equation (4) is the inductance matrix, where L ah and L b and L c and L af are the self-inductances of the stator windings A h , B, C, and A f respectively, and M j-k is the mutual inductance between the stator windings j and k (j ∈ {A h , B, C, A f}, k ∈ {A h , B, C, A f}).

[0025] Step 2 is specifically as follows:

[0026] Considering the influence of the disturbance voltage caused by the dead-time effect of the inverter, the model of the permanent magnet synchronous motor in the d-q axis system can be obtained from Equation (5): ​

[0027]

[0028] where u d-f and u q-f are the d-axis and q-axis disturbance voltages respectively. Transforming Equation (6) gives:

[0029]

[0030] Since the disturbance voltage caused by the dead zone is affected by various actual conditions and is difficult to obtain directly, this disturbance voltage can be used as the state variable of the system, and a disturbance observer can be designed to estimate the disturbance voltage. Its basic block diagram is as shown in Figure 2 Figure. Discretizing Equation (7) gives:

[0031]

[0032]

[0033] x1(k) = [i d (k) i q (k)] T and x2(k) = [u d-f (k) u q-f (k)] T (10)

[0034]

[0035]

[0036] In the design, the sampling period is very short, and it is considered that the disturbance voltage remains unchanged within one sampling period, that is:

[0037] x2(k) = x2(k + 1) (13)

[0038] Taking x2 as the observed object, the reduced-order disturbance observer is designed as:

[0039]

[0040] F is the gain matrix of the observer. To decouple the dq-axis components, F can be taken as F = kI 2×2

[0041] Then:

[0042]

[0043]

[0044] T s is the sampling period. To ensure the stability of the system The eigenvalues should satisfy:

[0045]

[0046] That is:

[0047] Step 3 is specifically as follows:

[0048] Inject a voltage vector with high-frequency rotation and constant amplitude, and superimpose this voltage vector on the original voltage. The high-frequency voltage is expressed as:

[0049]

[0050] where V i is the amplitude of the high-frequency voltage, and ω i is the frequency of the high-frequency voltage;

[0051] The high-frequency current response generated by the high-frequency voltage under a healthy motor is:

[0052]

[0053] In the formula

[0054] When a turn-to-turn short-circuit fault occurs, because the voltage frequency is very high, the stator resistance is ignored. From formula (2), we can obtain:

[0055] u a = pλ ah = p[L ah i a + M ah-b i b + M ah-c i c + μL ah i f (21)

[0056] The voltage equation in the short-circuit loop is:

[0057] r af i a +(r af + r f )i f = -pλ af = -pμλ ah (22)

[0058] where p is the differential operator, and λ ah and λ af are the magnetic flux linkage amplitudes of the healthy part and the faulty part of phase A respectively. Combining with formula (19), we get:

[0059]

[0060] When the turn - to - turn short - circuit fault is in its initial stage, r f >> r af , ignoring r af , the short - circuit current is in the opposite phase to the phase voltage, and the short - circuit current expression is:

[0061]

[0062] The high - frequency current response generated by the high - frequency voltage obtained by passing the current response in the stationary coordinate system through a band - pass filter is:

[0063]

[0064] Placing the current in a rotating coordinate system with a rotational speed of ω i , the fault characteristics can be transformed into a DC quantity and the remaining components are all high - frequency AC quantities. Using a low - pass filter to filter out the high - frequency components to obtain the fault characteristics for diagnosis:

[0065]

[0066] The beneficial effects of the present invention are:

[0067] The present invention is a method for improving the reliability of early turn - to - turn short - circuit fault diagnosis of a permanent - magnet synchronous motor. Compared with other fault diagnosis methods, by injecting high - frequency signals to amplify the fault characteristics and selecting the current response with high fault sensitivity as the fault characteristics, the purpose of accurately diagnosing the early turn - to - turn short - circuit fault is achieved. The unbalanced power supply of the inverter will generate negative - sequence current, which affects the accuracy of fault diagnosis. By using a voltage disturbance observer to eliminate the power - supply imbalance, the reliability of diagnosis is improved. Brief Description of the Drawings

[0068] Figure 1 is the equivalent circuit diagram of the turn - to - turn short - circuit fault adopted in the present invention;

[0069] Figure 2 is the structural block diagram of the voltage disturbance observer adopted in the present invention;

[0070] Figure 3 is the high - frequency signal injection block diagram based on the voltage disturbance observer in the present invention. Detailed Embodiment

[0071] The present invention will be described in detail below in conjunction with the drawings and specific embodiments.

[0072] A method for improving the reliability of early turn - to - turn short - circuit fault diagnosis of a permanent - magnet synchronous motor according to the present invention is specifically implemented according to the following steps:

[0073] Step 1: Establish a mathematical model of the motor under the condition of turn - to - turn short - circuit fault of PMSM, specifically:

[0074] When a turn - to - turn short - circuit fault occurs in the A - phase of the motor, the equivalent circuit of the turn - to - turn short - circuit fault is used, as Figure 1 shown, it can be known from Figure 1 that a short - circuit loop will be added to the A - phase winding of the motor. At this time, the resistance r f divides the A - phase into a healthy part and a faulty part, and the short - circuit turn ratio μ is defined as:

[0075]

[0076] where N f is the number of short - circuited turns of a certain - phase stator winding, and N is the total number of turns of a certain - phase stator winding;

[0077] According to the voltage equation of the faulty permanent - magnet synchronous motor, a mathematical model of the motor is established:

[0078]

[0079] where V abcf = [V ah V b V c V af T is the phase - voltage matrix of the stator winding. Among them, V ah , V b , V c , V af are the voltages of the healthy part of the A - phase, the voltages of the B - and C - phases, and the voltage of the faulty part of the A - phase respectively; i abcf = [i a i b i c i f T is the current matrix. i a , i b , i c , i f are the three - phase stator currents and the short - circuit current respectively; e m = [e ah e b e c e af T is the back - electromotive - force matrix of the three - phase stator winding and the short - circuit winding. Among them, e ah , e b , e c , e af are the back - electromotive forces of the healthy part of the A - phase, the back - electromotive forces of the B - and C - phases, and the back - electromotive force of the faulty part of the A - phase respectively;

[0080]

[0081] Equation (3) is the resistance matrix, where R​​​s is the stator resistance, r f is the resistance of the faulty winding A f ;

[0082]

[0083] Equation (4) is the inductance matrix, where L ah , L b , L c , L af are the self-inductances of the stator windings A h , B, C, A f respectively, and M j-k is the mutual inductance between the stator windings j and k (j ∈ {A h , B, C, A f}, k ∈ {A h , B, C, A f});

[0084] The voltage equation in the d-q axis system is obtained by coordinate transformation of Equation (2):

[0085]

[0086] where, u d , u q are the d-axis and q-axis voltages, i d , i q are the d-axis and q-axis currents, L d , L q are the d-axis and q-axis inductances, ω is the motor speed, θ r is the rotor position angle, and λ is the amplitude of the stator flux linkage.

[0087] Step 2: Design a voltage disturbance observer to compensate for the power supply imbalance caused by the compensation inverter according to the mathematical model of the inter-turn short circuit fault of the permanent magnet synchronous motor;

[0088] Specifically, Step 2 is as follows:

[0089] Considering the influence of the disturbance voltage caused by the dead-time effect of the inverter, the model of the permanent magnet synchronous motor in the d-q axis system can be obtained from Equation (5):

[0090]

[0091] where u d-f , u q-f are the d-axis and q-axis disturbance voltages respectively. Transforming Equation (6) gives:

[0092]

[0093] The disturbance voltage caused by the dead zone is affected by various actual conditions and is difficult to obtain directly. Therefore, this disturbance voltage can be used as the state variable of the system, and a disturbance observer can be designed to estimate the disturbance voltage. The basic block diagram is as shown in Figure 2 shown. Discretizing Equation (7) gives:

[0094]

[0095]

[0096] x1(k) = [i d (k) i q (k)] T and x2(k) = [u d-f (k) u q-f (k)] T (10)

[0097]

[0098]

[0099] In the design, the sampling period is very short, and it is considered that the disturbance voltage remains unchanged within one sampling period, that is:

[0100] x2(k) = x2(k + 1) (13)

[0101] Taking x2 as the observed object, a reduced-order disturbance observer is designed as:

[0102]

[0103] F is the gain matrix of the observer. To decouple the dq-axis components, F can be taken as F = kI 2×2

[0104] Then:

[0105]

[0106]

[0107] T s is the sampling period. To ensure the stability of the system the eigenvalues should satisfy:

[0108]

[0109] That is:

[0110] Step 3: After compensation, extract the high-frequency current response under the high-frequency signal injection of the permanent magnet synchronous motor through a band-pass filter, and then perform coordinate transformation and low-pass filter to extract the fault characteristics.

[0111] Step 3 is specifically as follows:

[0112] Inject a voltage vector with high-frequency rotation and constant amplitude, and superimpose this voltage vector on the original voltage. This high-frequency voltage is expressed as:

[0113]

[0114] where V i is the amplitude of the high-frequency voltage, and ω i is the frequency of the high-frequency voltage;

[0115] The high-frequency current response generated by the high-frequency voltage under a healthy motor is:

[0116]

[0117] In the formula

[0118] When a turn-to-turn short circuit fault occurs, because the voltage frequency is very high, the stator resistance is ignored, and the following can be obtained from formula (2):

[0119] u a = pλ ah = p[L ah i a + M ah-b i b + M ah-c i c + μL ah i f (21)

[0120] The voltage equation in the short-circuit loop is:

[0121] r af i a +(r af + r f )i f = -pλ af = -pμλ ah (22)

[0122] where p is the differential operator, and λ ah , λ af are the magnetic flux linkage amplitudes of the healthy part and the faulty part of phase A respectively. Combining with formula (19), we get:

[0123]

[0124] When the turn-to-turn short circuit fault is in the initial stage, r f >> r af , ignoring r af , the short-circuit current is in the opposite phase to the phase voltage, and the short-circuit current expression is:

[0125]

[0126] The high-frequency current response generated by the high-frequency voltage obtained by passing the current response in the stationary coordinate system through a band-pass filter is as follows:

[0127]

[0128] Placing the current in a rotating coordinate system with a rotational speed of ω i can transform the fault characteristics into a direct current quantity and the remaining components are all high-frequency alternating current quantities. A low-pass filter is used to filter out the high-frequency components to obtain the fault characteristics for diagnosis:

[0129]

[0130] The control block diagram of the system for improving the reliability of early inter-turn fault diagnosis of a permanent magnet synchronous motor is as shown in Figure 3 . Based on vector control, the system uses u d-f and u q-f as the observation objects, and obtains the estimated disturbance voltage value Figure 2 through a voltage disturbance observer (such as ). The feedback compensates u d and u q to avoid unbalanced inverter power supply. By injecting high-frequency signals with variable amplitudes and frequencies in the αβ coordinate system, and using three Hall current sensors to detect the three-phase currents in the three-phase stationary coordinate system. At this time, the three-phase currents include fundamental frequency currents and high-frequency currents. The stationary three-phase currents are transformed into the currents i α and i β in the two-phase stationary coordinate system through Clark transformation. The high-frequency response currents i αh and i βh are obtained through band-pass filter processing. The high-frequency currents pass through a similar Park coordinate transformation with an angle of ω i t to obtain the positive sequence, negative sequence components and fault components of the current. At this time, only the fault component is a direct current quantity. The alternating current components are filtered out through a low-pass filter to obtain the direct current component for fault diagnosis. A method for improving the reliability of early inter-turn fault diagnosis of a permanent magnet synchronous motor according to the present invention compensates the unbalanced inverter power supply problem through a voltage disturbance observer, avoids the influence of inverter nonlinearity on the fault characteristics, and improves the reliability of fault diagnosis. The fault characteristics are amplified by injecting high-frequency voltage signals, and early diagnosis of inter-turn short circuit faults is realized by extracting the direct current fault characteristic current response with high fault sensitivity.

Claims

1. A method for improving the reliability of early turn - to - turn short - circuit fault diagnosis of permanent magnet synchronous motors, characterized in that, The implementation is carried out in the following steps: Step 1: Establish a mathematical model for the inter-turn short-circuit fault of the permanent magnet synchronous motor; The specific content of Step 1 is as follows: When a turn-to-turn short circuit fault occurs in the A-phase of the motor, a short circuit loop will be added to the A-phase winding of the motor, and at this time the resistance r f divides the A-phase into a healthy part and a faulty part, and defines the short circuit turn ratio μ as: where N f is the number of short - circuited turns of a certain - phase stator winding, and N is the total number of turns of a certain - phase stator winding; Establish the motor mathematical model according to the fault motor voltage equation of the permanent magnet synchronous motor: where V abcf =[V ah V b V c V af T is the phase voltage matrix of the stator winding, where V ah , V b , V c , V af are the healthy part voltage of phase A, the voltages of phases B and C, and the faulty part voltage of phase A respectively; i abcf =[i a i b i c i f T is the current matrix, where i a , i b , i c , i f are the three-phase stator currents and the short-circuit current respectively; e m =[e ah e b e c e af T is the back electromotive force matrix of the three-phase stator winding and the short-circuit winding, where e ah , e b , e c , e af are the back electromotive force of the healthy part of phase A, the back electromotive forces of phases B and C, and the back electromotive force of the faulty part of phase A respectively;​​​ The voltage equation in the d-q axis system is obtained by coordinate transformation of Formula (2): where, u d , u q are the d - axis and q - axis voltages, i d , i q are the d - axis and q - axis currents, L d , L q are the d - axis and q - axis inductances, ω is the motor speed, θ r is the rotor position angle, and λ is the amplitude of the stator flux linkage; In Step 1: Equation (3) is a resistance matrix, where R s is the stator resistance, r f is the resistance of the faulty winding A f ; Equation (4) is the inductance matrix, where L ah , L b , L c , L af are the self-inductances of the stator windings A h , B, C, A f respectively, and M j-k is the mutual inductance between the stator windings j and k (j ∈ {A h , B, C, A f}, k ∈ {A h , B, C, A f}); Step 2: Design a voltage disturbance observer to compensate for the power supply imbalance caused by the compensation inverter according to the mathematical model of the inter-turn short-circuit fault of the permanent magnet synchronous motor; The specific content of Step 2 is as follows: The model of the permanent magnet synchronous motor in the d-q axis system can be obtained from Equation (5): where u d-f and u q-f are the d-axis and q-axis disturbance voltages respectively. Transforming Equation (6) gives: Design a disturbance observer to estimate the disturbance voltage. The discretization of Equation (7) can obtain: x1(k) = [i d (k) i q (k)] T 、x2(k) = [u d-f (k) u q-f (k)] T (10) In the design, the sampling period is very short. It is considered that the disturbance voltage remains unchanged within one sampling period, that is: x2(k) = x2(k + 1) (13) Taking x2 as the observed object, design a reduced-order disturbance observer as: F is the gain matrix of the observer. To decouple the dq-axis components, F can be taken as F = kI 2×2 Then: T s is the sampling period. To ensure the stability of the system the eigenvalues of should satisfy: That is: Step 3: After compensation, extract the high-frequency current response under the injection of the high-frequency signal of the permanent magnet synchronous motor through a band-pass filter, and then perform coordinate transformation and low-pass filter to extract the fault characteristics.

2. The method for improving the reliability of early turn-to-turn short circuit fault diagnosis of a permanent magnet synchronous motor according to claim 1, characterized in that, The specific content of Step 3 is as follows: Inject a high-frequency rotating voltage vector with a constant amplitude, and superimpose this voltage vector on the original voltage. This high-frequency voltage is expressed as: Among them, V i is the high-frequency voltage amplitude, and ω i is the frequency of the high-frequency voltage; The high-frequency current response generated by the high-frequency voltage under a healthy motor is: wherein When an inter-turn short-circuit fault occurs, because the voltage frequency is very high, the stator resistance is ignored and Formula (2) can be used to obtain: u a = pλ ah = p[L ah i a + M ah-b i b + M ah-c i c + μL ah i f (21) The voltage equation in the short-circuit loop is: r af i a +(r af +r f )i f =-pλ af =-pμλ ah (22) where p is the differential operator, λ ah , λ af are the magnetic flux linkage amplitudes of the healthy and faulty parts of phase A respectively. Combining with Equation (19), we get: When the turn-to-turn short-circuit fault is in its initial stage, r f >> r af , ignoring r af , the short-circuit current is opposite in phase to the phase voltage, and the expression for the short-circuit current is: The high-frequency current response generated by the high-frequency voltage obtained by passing the current response in the stationary coordinate system through a band-pass filter is: Place the current in a rotating coordinate system with a rotational speed of ω i and after processing with a low-pass filter: The fault characteristics are transformed into a direct current quantity, and the remaining components are all high-frequency alternating current quantities. Use a low-pass filter to filter out the high-frequency components to obtain the fault characteristics for diagnosis.

Images