Fault diagnosis method for switch tubes in open-winding permanent magnet synchronous motor dual inverter system

Abstract

The present invention relates to a method for diagnosing faults of switch tubes in a dual inverter system of an open-winding permanent magnet synchronous motor, comprising the following steps: step 1, calculating the average value of the phase current after standardization and the output average value of a PR controller, judging the corresponding flag bit symbol according to a threshold range, and determining the combination of the fault phase and the switch tube with a fault by using judgment logic; step 2, blocking the switch tube connected to the fault phase to obtain the direction of the reverse electromotive force of the fault; using a current Hall sensor for detection; if the current Hall sensor detects the current, the switch tube is normal, and if the current is not detected, the switch tube has an open circuit fault. The present invention is aimed at a three-phase open-winding permanent magnet synchronous motor, and not only uses a zero-sequence controller unique to the three-phase open-winding permanent magnet synchronous motor, but also uses the back electromotive force for fault location, with better performance.

Description

Technical Field

[0001] The present invention belongs to the technical field of synchronous motor drive, and in particular relates to a method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor. Background Art

[0002] Fault diagnosis of the drive system is the guarantee of the normal operation of the system. Only by accurately determining the location of the fault can fault-tolerant control be performed to ensure that the system continues to operate in the event of a fault. Fault diagnosis methods can generally be divided into model-free fault diagnosis and model-based fault diagnosis. The former usually requires complex algorithms such as machine learning, fuzzy algorithms, and other artificial intelligence, which require a large amount of processor resources and are difficult to implement in engineering. Model-based fault diagnosis usually analyzes the obtained voltage and current signals according to the system model to diagnose the system fault. Among them, the frequency domain analysis method is relatively accurate, but the amount of calculation is large and the calculation time is long. The time domain analysis method is simple to implement and has a small amount of calculation, and is currently widely used.

[0003] According to the different processed signals, the diagnostic methods based on processed signals can be divided into voltage type and current type. The fault diagnosis method using voltage signals has the advantage of fast diagnostic speed, but it requires additional hardware circuits such as voltage sampling circuits and comparison circuits. In the open-winding permanent magnet synchronous motor drive system, due to the increase in the number of switching tubes, the volume and cost of the sampling circuit of the voltage-based fault diagnosis method are significantly increased, and it is also easily affected by the switching frequency, drive circuit, etc. Compared with the voltage-type diagnostic method, the current-type diagnostic method does not require the addition of additional voltage sampling circuits, which is more conducive to implementation in engineering. For traditional neutral point connected permanent magnet synchronous motors, common current diagnostic methods include the average current Park method and the current vector trajectory method.

[0004] For open-winding motor drive systems, due to the symmetry of the structure, different faults will generate the same fault current after a fault occurs, so many fault diagnosis methods cannot be directly applied to open-winding motor drive systems. It is impossible to achieve accurate switch tube fault diagnosis without adding hardware detection circuits. Summary of the invention

[0005] In order to solve the problems of the prior art, the present invention further proposes a method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor.

[0006] The present invention relates to a method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor, comprising the following steps:

[0007] Step 1: Calculate the average value of the phase current after standardization and the output average value of the PR controller, determine the corresponding flag bit sign according to the threshold range, and use the judgment logic to determine the combination of the faulty phase and the faulty switch tube;

[0008] Step 2: Block the switch connected to the fault phase to obtain the direction of the opposite electromotive force of the fault; give the switch to be tested a pulse width modulated drive signal so that it can form a loop with the anti-parallel diode and winding of the non-faulty switch, and use the current Hall sensor for detection; if the current Hall sensor detects current, the switch is normal; if no current is detected, the switch has an open circuit fault.

[0009] In step 1, the average current after standardization is calculated as follows:

[0010] Define the current vector magnitude as

[0011]

[0012] In the formula, —current vector magnitude; i d —d-axis current; i q —q-axis current; divide the current value of each phase by the current vector modulus to calculate the current value of each phase after standardization

[0013]

[0014] In the formula, i xN (n)—per unit current value; i x (n)—x phase current value; According to the above formula, the maximum current value can be obtained

[0015]

[0016] In the formula, I m —phase current amplitude; calculate the average current value of each phase after standardization:

[0017]

[0018] In the formula, <|i xN |>—average current value after standardization; ω—electrical angular velocity;

[0019] The output average value of the PR controller is calculated as:

[0020] Under normal working conditions, the scoring period is 2π / ω, then <|i xN The value of |> is 0.6366; so the fault diagnosis variable e is defined x

[0021] e x =0.6366-<|ixN |> (5)

[0022] Under normal working conditions, x The value of is close to 0; when a phase fails, e x The value of will change significantly; at the same time, the output of PR is used to make fault judgment and obtain the average value of the output of PR controller.

[0023] In step 1, according to e x and the average value of u0 to obtain the fault characteristics; first define the diagnostic flag

[0024]

[0025]

[0026] Where N, P1, P2, P3, L and H are all defined diagnostic flags. 0av and e x The value range can be used to determine E x Whether the flag bit of and U is valid, the three phases A, B, and C are all fault phases, symmetrical and 120° apart, and then the fault phase and fault combination that may have a fault are determined according to the following logic: Phase A combination 1: P2 NN

[0027] H

[0028]

[0029] If the flag bit E a =P2,E b =N,E c =N,U=H; at this time, it can be determined that the possible fault combination in phase A is {S 11 ,S 22}; At this time, S 11 and S 22 Faults may exist individually or simultaneously.

[0030] If the flag bit E a =P2,E b =N,E c =N,U=L, it can be determined that the possible fault combination in phase A is {S 12 , S 21}; At this time, S 12 and S 21 Faults may exist individually or simultaneously.

[0031] If the flag bit E a =P2,E b =N,E c=N, U = H / L, the average value of the PR controller appears positive and negative in one cycle. At this time, it can be determined that the possible fault combination in phase A is {{S 11 ,S 22},{S 12 ,S 21}}, it may be that three of the four switching tubes are faulty, or the two upper tubes or the two lower tubes are faulty, or all of them may be faulty.

[0032] In step 2, if phase A is the fault phase, the back electromotive force of phase A can be obtained according to formula (8):

[0033]

[0034] Where, M is the mutual inductance between phases; —A phase flux; e af —A phase winding voltage.

[0035] In step 2, when the back electromotive force on the fault phase A is positive, it is divided into the following three cases:

[0036] (1) Give S 11 A pulse width modulation drive signal generates a positive current on phase A. This current is detected by the controller, and the S 11 No failure occurred, S 22 Failure occurred; turn off S 11 , generate 5ms PWM signal to S 22 ; If the controller does not detect a positive current, it determines S 11 Normal, S 22 Fault;

[0037] (2) Give S 11 A PWM drive signal, A phase does not generate a positive current, which means S 11 If a fault occurs, continue to give S 22 The same PWM signal, at this time no current is generated on phase A, which finally proves that S 11 Fault, S 22 normal;

[0038] (3) Give S 11 A PWM drive signal, no forward current is detected on phase A, then it is determined that S 11 Failure, continue to give S 22 For the same PWM signal, if there is no current on phase A, then it is determined that S 11 and S 22 All failed.

[0039] In step 2, when the fault back electromotive force is positive, the PI structure is used to generate a PWM drive signal, and the drive signal is given to the switch tube S 11 ;

[0040] Step 2.1.1: If no forward current is generated at this time, it can be determined that the switch tube S 11 Fault occurs, then switch S 11 The drive signal is blocked and the signal is given to the switch tube S 22 If a forward current is generated at this time, the switch tube S 22 No fault has occurred; if no current is generated, S 22 There was also a malfunction;

[0041] Step 2.1.2: If a forward current is generated at this time, it can be determined that the switch tube S 11 If no fault occurs, the switch tube S 22 A fault may occur; turn on the switch S 11 The drive signal is blocked and the signal is given to the switch tube S 22 If no forward current is generated at this time, the switch tube S 22 A fault occurs; if a forward current is generated, then step 1 is judged wrong, and the switch tube S 11 and S 22 No fault occurred; return to step 1 to continue judgment; if both times the circuit is generated, then the switch tube S 11 and S 22 All faults,

[0042] In step 2, when the fault back electromotive force is negative, the PI structure is used to generate a PWM drive signal, and the drive signal is given to the switch tube S 12 ;

[0043] Step 2.2.1: If no negative current is generated at this time, it can be determined that the switch tube S 12 Fault occurs, then switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If a negative current is generated at this time, the switch tube S 21 No fault has occurred; if no current is generated, S 21 A malfunction occurs;

[0044] Step 2.2.2: If a negative current is generated at this time, it can be determined that the switch tube S 12 If no fault occurs, the switch tube S 21 A fault may occur; next, switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If no negative current is generated at this time, the switch tube S 21Fault occurs; if negative current is generated, then step 1 is wrong, and the switch tube S 12 and S 21 If no fault occurs, return to step 1 to continue the judgment.

[0045] Beneficial Effects

[0046] The present invention is aimed at a three-phase open-winding permanent magnet synchronous motor. Not only does it use a zero-sequence controller unique to the three-phase open-winding permanent magnet synchronous motor in the first step, but it also uses back electromotive force to locate faults, thereby achieving better performance.

[0047] The present invention uses the output of the PR controller to determine the fault type and fault phase, and uses the back electromotive force to determine the position of the fault switch tube, thereby solving the problem that the traditional fault diagnosis technology cannot accurately obtain the position of the fault switch tube due to the symmetry of the dual inverter topology, ensuring that the fault-tolerant control scheme is put into use in a timely and accurate manner, and improving system reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a structural diagram of a common DC bus open-winding permanent magnet synchronous motor used in the present invention;

[0049] Figure 2 It is a circuit structure analysis diagram of the dual inverter structure of the present invention;

[0050] Figure 3 It is a synthetic diagram of the effective voltage vector of the dual inverter used in the present invention;

[0051] Figure 4 It is a control block diagram of the open-winding permanent magnet synchronous motor system of the present invention;

[0052] Figure 5 It is the current path diagram after the A phase S11 fault in the present invention;

[0053] Figure 6 This is a diagram of the diagnosis process of the switch tube S11 in the present invention;

[0054] Figure 7 This is a diagram of the fault diagnosis process of the switch tubes S11 and S12 in the present invention;

[0055] Figure 8 This is a diagram of the fault diagnosis process of the switch tubes S11 and S21 in the present invention;

[0056] Fig. 9 It is a flowchart of the first step of the fault diagnosis described in the present invention;

[0057] Fig.10 It is the flow chart of the second step of the fault diagnosis described in the present invention. DETAILED DESCRIPTION

[0058] The following combination Figures 1 to 10 This embodiment will be described in detail.

[0059] The method of the present invention is used for fault diagnosis of open-winding permanent magnet synchronous motor drive system with common DC bus, and the application topology is as follows: Figure 1 shown.

[0060] The motor is a traditional permanent magnet synchronous motor with an open neutral point. It is driven by a dual inverter structure, with the two inverters being INV1 and INV2. The DC sides of the two inverters are powered by the same DC bus. Due to the existence of the same DC bus, there is a zero-sequence current in the system, which forms a loop through the DC bus and the three-phase winding, thus generating additional system losses and six-order torque pulsations. In order to suppress the influence of the zero-sequence current difference, a proportional resonant (PR) controller is used to suppress the zero-sequence current.

[0061] For a certain phase, there are four switch tubes (S 11 , S 12 , S 21 and S 22 ) is connected to the positive and negative busbars and powered by two independent bridge arms, such as Figure 2 shown.

[0062] A dead time is added between the two switch tubes on the same bridge arm to ensure that the upper and lower tubes cannot be turned on at the same time to cause a short circuit fault. The output voltage of each bridge arm is: a1 、u b1 、u c1 、u a2 、u b2 and u c2 , the voltage acting on each phase is the voltage difference between the midpoints of the two bridge arms, as shown in formula (1).

[0063]

[0064] In the formula, u x —x-phase winding voltage; u x1 —Inverter 1x phase voltage; u x2 —Inverter 2x phase voltage.

[0065] like Figure 3 As shown in the figure, it is the vector synthesis scheme used by the dual inverters. Each inverter adopts the space vector modulation strategy (SVPWM) and can output 8 voltage vectors (V1-V8). Finally, 64 voltage vectors can be synthesized. Figure 3 In the figure, the dotted middle vector is used to synthesize the final reference voltage acting on the OEW-PMSM, so that the zero-sequence voltage generated by the effective voltage vector synthesized by the two inverters is zero.

[0066] The current closed-loop control block diagram is as follows Figure 4 As shown, it includes the traditional speed outer loop and current inner loop. It includes PI regulator, comparator, Clark and Park transformation, three-phase current sampling, encoder and other modules. The zero-sequence component in the system is shown in formula (2) and formula (3).

[0067]

[0068]

[0069] In the formula, i0 is zero-sequence current; u0 is zero-sequence voltage; i x —x phase current; u x —x phase voltage.

[0070] A proportional resonant controller is used to suppress the zero-sequence current in the common bus dual inverter structure. The input reference current of the proportional resonant controller is zero, and the feedback quantity is shown in formula (2). The transfer function of the proportional resonant controller is shown in formula (4).

[0071]

[0072] Where, G(s) is the proportional resonant controller transfer function; K p —Proportional coefficient; K r —Resonance coefficient; ω c —cut-off frequency; ω0—resonant frequency.

[0073] The output of the proportional resonant controller is u0. After u0 enters the SVPWM module, the time of the zero voltage vector that each inverter needs to compensate is calculated according to the size of u0 to achieve the purpose of zero-sequence current suppression.

[0074] The working process is:

[0075] The encoder obtains the motor rotor position and speed. After comparing the actual speed with the given speed, it enters the speed outer loop PI controller and outputs the q-axis reference current. The three current sensors obtain the three-phase current, which is converted into the dq-axis current I after Clark-Park transformation. d ,I q After comparing with the given reference current, the output U d and U q .U d and U q Distribute it equally to the two inverters, and then perform SVPWM after inverse Park transformation. Then, according to the size of u0, redistribute the zero vector size of the two inverters. Finally, output the drive signal to each switch tube.

[0076] Assume that after A fails, the average current value in the system changes. The average current value after standardization can be used to determine whether there is a fault in the system. At the same time, through analysis, it can be known that when S 11 After a fault occurs, the zero-sequence current frequency changes from three times the original frequency to the fundamental frequency, so the frequency output by the proportional resonant controller also changes from three times the original frequency to the fundamental frequency. Ultimately, these two characteristics are used to determine the phase where the fault occurs and the combination of faulty switch tubes.

[0077] For phase A, if the current of phase A is positive, the switch S 11 and S 22 Conductivity, such as Figure 5 Assume that the switch tube S of phase A 11 At some point a fault occurs. The phase current changes to flow through S 12 The anti-parallel diode and S 22 , like Figure 5 (a) as shown. 22 Failure and S 11 and S 22 At the same time, the fault current path is Figure 5 (b) and 5(c).

[0078] After a phase fails, it can be divided into three fault combinations:

[0079] Combination 1: {S 11 ,S 22};

[0080] Combination 2: {S 12 ,S 21};

[0081] Combination 3: {{S 11 ,S 22},{S 12 ,S 21}}.

[0082] One of the switch tubes in the brackets may fail, or both of them may fail. The third type is a combination of the first and second types of failures.

[0083] The following are the diagnostic steps:

[0084] The method of the present invention is mainly divided into two steps. Step 1 is to determine which phase has which fault combination, and then perform different step 2 according to different fault switch tube combinations; step 2 is to first block the four switch tubes connected to the fault phase, and obtain the direction of the back electromotive force. Assume S 11 When a fault occurs, when the direction of the back electromotive force is positive, S 11 A driving signal, no current is generated.22 When the driving signal is high, the current of phase A rises and the current flows through S 22 and S 12 When the drive signal is low, the current of phase A decreases and the current flows through S 12 and S 21 anti-parallel diode.

[0085] Step 1: Calculate the average value of the phase current after standardization and the output average value of the PR controller, then determine the corresponding flag bit sign according to the threshold range, and finally use the judgment logic to determine the combination of the faulty phase and the faulty switch tube.

[0086] Step 1.1: Use the current average value to determine whether there is a fault in the system and define the current vector modulus as

[0087]

[0088] In the formula, —current vector magnitude; i d —d-axis current; i q —q-axis current.

[0089] Then divide the current value of each phase by the current vector modulus to calculate the current value of each phase after standardization.

[0090]

[0091] In the formula, i xN (n)—per unit current value; i x (n)—x-phase current value.

[0092] According to the above formula, the maximum current value can be obtained

[0093]

[0094] In the formula, I m —Phase current amplitude.

[0095] The average current value of each phase after standardization is calculated as

[0096]

[0097] In the formula, <|i xN |>—average current value after standardization; ω—electrical angular velocity.

[0098] Under normal working conditions, the scoring period is 2π / ω, then <|i xN The value of |> is 0.6366. So we define the fault diagnosis variable e x

[0099] ex =0.6366-<|i xN |> (9)

[0100] From formula (5), we can see that under normal working conditions, e x The value of is close to 0. When a phase fails, e x The value of will change significantly. However, only e x Intelligent judgment of which fault is present cannot completely determine the specific fault, so it is necessary to use the output of PR to make a judgment. The PR controller generates the zero-sequence current generated by the dead zone and the third back electromotive force in the zero-sequence voltage compensation system, so the output of the PR controller is the same as the zero-sequence current waveform, and the frequency is also three times the fundamental frequency. Therefore, the average value of the output of the controller is used to obtain the average value of the output of the PR controller. According to e x The average value of and u0 is used to obtain the fault characteristics. First, define the diagnostic flag:

[0101]

[0102]

[0103] Where N, P1, P2, P3, L and H are all defined diagnostic flags. 0av and e x The value range can be used to determine E x The flag bits of and U are valid, and then the fault phase and fault combination that may have a fault are determined according to the following logic:

[0104]

[0105] Step 1.2: Determine the faulty phase and the combination of faulty switch tubes according to the judgment logic, and proceed to the second step after confirmation. The following takes phase A as an example for explanation.

[0106] Step 1.2.1: If the flag is E a =P2,E b =N,E c =N,U=H. At this time, it can be determined that the possible fault combination in phase A is {S 11 ,S 22}. At this time, it may be S 11 and S 22 There may be a single fault or a combination of faults. Next, proceed to step 2.1.

[0107] Step 1.2.2: If the flag is E a =P2,E b =N,E c=N,U=L. At this time, it can be determined that the possible fault combination in phase A is {S 12 ,S 21}. At this time, it may be S 12 and S 21 There may be a single fault or a combination of faults. Next, proceed to step 2.1.

[0108] Step 1.2.3: If the flag is E a =P2,E b =N,E c =N,U=H / L. Here, the average value of the PR controller appears positive and negative in one cycle. At this time, it can be determined that the possible fault combination in phase A is {{S 11 ,S 22},{S 12 ,S 21}}. At this time, it may be that three of the four switch tubes are faulty, or two upper tubes or two lower tubes are faulty, or both are faulty. It is necessary to use step 2 for further judgment. Next, go to step 2.1.

[0109] Step 2: After the fault phase and fault combination are determined in step 1, the four switch tubes (S 11 ,S 12 ,S 21 and S 22 ) is completely blocked. If phase A is assumed to be the fault phase, the reverse electromotive force of phase A can be obtained according to formula (12).

[0110]

[0111] Where, M is the mutual inductance between phases; —A phase flux; e af —A phase winding voltage.

[0112] When the back electromotive force on phase A is positive, the situations are divided into the following categories:

[0113] (1) Give S 11 A pulse width modulation (PWM) drive signal lasts for a period of time. During this period of time, a positive current is generated on phase A. This current is detected by the controller, and it can be determined that S 11 No failure occurred, S 22 A failure occurred. To further confirm that S 22 Is it normal? Close S 11 , generates a 5ms PWM signal to S 22 If the controller does not detect a positive current, it determines that S 11 Normal, S 22Fault.

[0114] (2) Give S 11 A PWM drive signal, during this period of time, no forward current is generated on phase A, which proves that S 11 Failure occurs, then continue to give S 22 The same PWM signal, at this time, no current is generated on phase A, which can finally prove that S 11 Fault, S 22 normal.

[0115] (3) Give S 11 A PWM drive signal, during this period of time, no forward current is detected on phase A, then it can be determined that S 11 Failure, then continue to give S 22 For the same PWM signal, if there is no current on phase A, it can be determined that S 11 and S 22 All failed.

[0116] The PI controller is used for closed-loop control. The reference current is 0.1A and the feedback current is the actual fault phase current. The selection of the DC reference current should be determined according to the specific parameters of the system. If the parameter is large, additional torque pulsation may be generated; if the parameter is small, it may be affected by the sampling accuracy and interference, resulting in misdiagnosis. The overall diagnostic process is as follows: Figure 6 and Figure 7 As shown, the specific steps are as follows:

[0117] Step 2.1: Block the four switch tubes connected to the fault phase. When the fault phase back electromotive force is positive, use the PI structure to generate a PWM drive signal and send the drive signal to the switch tube S. 11 .

[0118] Step 2.1.1: If no forward current is generated at this time, it can be determined that the switch tube S 11 Fault occurs, then switch S 11 The drive signal is blocked and the signal is given to the switch tube S 22 If a forward current is generated at this time, the switch tube S 22 No fault has occurred, the waveform of the diagnostic process is as follows Figure 6 If no current is generated, S 22 There was also a malfunction;

[0119] Step 2.1.2: If a forward current is generated at this time, it can be determined that the switch tube S 11 If no fault occurs, the switch tube S 22 A fault may occur. Next, switch S 11 The drive signal is blocked and the signal is given to the switch tube S22 If no forward current is generated at this time, the switch tube S 22 When a fault occurs, the waveform of the diagnostic process is as follows: Figure 7 If a forward current is generated, then the judgment in step 1 is wrong, and the switch tube S 11 and S 22 No fault occurred. Return to step 1 to continue the judgment. If both times the circuit is generated, then the switch tube S 11 and S 22 All faults, the waveform of the diagnosis process is as follows Figure 8 shown.

[0120] Step 2.2: Block the four switch tubes connected to the fault phase. When the fault back electromotive force is negative, use the PI structure to generate a PWM drive signal and send the drive signal to the switch tube S 12 .

[0121] Step 2.2.1: If no negative current is generated at this time, it can be determined that the switch tube S 12 Fault occurs, then switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If a negative current is generated at this time, the switch tube S 21 No fault has occurred. If no current is generated, S 21 A malfunction occurs;

[0122] Step 2.2.2: If a negative current is generated at this time, it can be determined that the switch tube S 12 If no fault occurs, the switch tube S 21 A fault may occur. Next, switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If no negative current is generated at this time, the switch tube S 21 If a negative current is generated, then step 1 is wrong, and the switch tube S 12 and S 21 No fault has occurred. Return to step 1 to continue the judgment.

[0123] Step 2.3: Block the four switches connected to the fault phase. When the reverse electromotive force of the fault is positive, use Figure 6 The structure shown generates a PWM drive signal and sends the drive signal to the switch tube S 11 .

[0124] Step 2.3.1: If no forward current is generated at this time, it can be determined that the switch tube S 11 Fault occurs, then switch S 11 The drive signal is blocked and the signal is given to the switch tube S 22If a forward current is generated at this time, the switch tube S 22 No fault has occurred. If no current is generated, S 22 A malfunction occurs;

[0125] Step 2.3.2: If a forward current is generated at this time, it can be determined that the switch tube S 11 If no fault occurs, the switch tube S 22 A fault may occur. Next, switch S 11 The drive signal is blocked and the signal is given to the switch tube S 22 If no forward current is generated at this time, the switch tube S 22 If a forward current is generated, then step 1 is wrong, and the switch tube S 11 and S 22 No failures occurred.

[0126] Step 2.3.3: When the fault back electromotive force is negative, use the PI structure to generate a PWM drive signal and give the drive signal to the switch tube S. 12 .

[0127] Step 2.3.4: If no negative current is generated at this time, it can be determined that the switch tube S 12 Fault occurs, then switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If a negative current is generated at this time, the switch tube S 21 No fault has occurred. If no current is generated, S 21 A fault also occurs. The diagnostic process is as follows Figure 7 As shown;

[0128] Step 2.3.5: If a negative current is generated at this time, it can be determined that the switch tube S 12 If no fault occurs, the switch tube S 21 A fault may occur. Next, switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If no negative current is generated at this time, the switch tube S 21 If a negative current is generated, then step 1 is wrong, and the switch tube S 12 and S 21 No faults occurred. 11 , S 12 , S 21 and S 22 If no fault occurs, return to step 1 to continue the judgment.

[0129] In step 2, step 2.3 is a combination of step 2.1 and step 2.2, which is to execute step 2.1 and step 2.2 respectively, and then determine the position of the faulty switch tube according to the results of 2.1 and 2.2.

[0130] The above contents are only preferred implementation cases of the present invention and are not intended to limit the implementation scheme of the present invention. Ordinary technicians in this field can easily make corresponding changes or modifications based on the main concepts and spirits of the present invention. For example, the number of indoor standard specimens and the number of cores taken at the construction site can be increased or decreased according to the accuracy requirements. Therefore, the protection scope of the present invention shall be based on the protection scope required by the claims.

Claims

1. A method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor, characterized in that: The steps include: Step 1: Calculate the average value of the phase current after standardization and the output average value of the PR controller, determine the corresponding flag bit sign according to the threshold range, and use the judgment logic to determine the combination of the faulty phase and the faulty switch tube; under normal working conditions, the scoring period is 2π / ω, then <|i xN The value of |> is 0.6366; so the fault diagnosis variable e is defined x yes x =0.6366-<|i xN |> (5) <|i xN |>—average current value after standardization; Under normal working conditions, x The value of is close to 0; When a phase fails, x The value of will change significantly; at the same time, the output of PR is used to judge the fault and obtain the average value of the output of PR controller; According to e x and the average value of the zero-sequence controller output value u0 to obtain the fault characteristics; first define the diagnostic flag Where th1 and th2 are E x The threshold value has different values ​​in different working environments; N, P1, P2, P3, L and H are all diagnostic flags. x The range of values ​​can be used to determine E x Whether the flag bit of and U is valid, the three phases A, B, and C are all fault phases, symmetrical and 120° apart. The fault phase and fault combination that may exist are determined according to the following logic: A phase combination 1: P2 NNH A phase combination 2: P2 NNL A phase combination 3: P3 NNH / N B phase combination 1: N P2 NH B phase combination 2: N P2 NL B phase combination 3: N P3 NH / N C phase combination 1: NN P2 H C phase combination 2: NN P2 L C phase combination 3: NN P3 H / N; Step 2: Block the switch connected to the fault phase to obtain the direction of the opposite electromotive force of the fault; give the switch to be tested a pulse width modulated drive signal so that it can form a loop with the anti-parallel diode and winding of the non-faulty switch, and use the current Hall sensor for detection; if the current Hall sensor detects current, the switch is normal; if no current is detected, the switch has an open circuit fault.

2. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 1, characterized in that: In step 1, the average current after standardization is calculated as follows: Define the current vector magnitude as In the formula, —current vector magnitude; i d —d-axis current; i q —q-axis current; divide the current value of each phase by the current vector modulus to calculate the current value of each phase after standardization In the formula, i xN (n)—per unit current value; i x (n)—x phase current value; According to the above formula, the maximum current value can be obtained In the formula, I m —phase current amplitude; The average current value of each phase after standardization is calculated as In the formula, <|i xN |>—average current value after standardization; ω—electrical angular velocity.

3. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 1, characterized in that: There are four switch tubes (S11, S12, S21 and S22) in the fault phase A. If the flag is E a =P2,E b =N,E c =N,U=H; at this time, it can be determined that the possible fault combination in phase A is {S 11 ,S 22 }; At this time, S 11 and S 22 Faults may exist individually or simultaneously.

4. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 1, characterized in that: There are four switch tubes (S11, S12, S21 and S22) in the fault phase A. If the flag is E a =P2,E b =N,E c =N,U=L, it can be determined that the possible fault combination in phase A is {S 12 ,S 21 }; At this time, S 12 and S 21 Faults may exist individually or simultaneously.

5. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 1, characterized in that: There are four switch tubes (S11, S12, S21 and S22) in the fault phase A. If the flag is E a =P2,E b =N,E c =N, U = H / L, the average value of the PR controller appears positive and negative in one cycle. At this time, it can be determined that the possible fault combination in phase A is {{S 11 ,S 22 },{S 12 ,S 21 }}, it may be that three of the four switching tubes are faulty, or the two upper tubes or the two lower tubes are faulty, or there may be faults at the same time.

6. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 1, characterized in that: In step 2, if phase A is the fault phase, the back electromotive force of phase A can be obtained according to formula (8): Where, M is the mutual inductance between phases; —A phase flux; e af —A phase winding voltage.

7. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 6, characterized in that: In step 2, when the back electromotive force on the fault phase A is positive, it is divided into the following three cases: (1) Give S 11 A pulse width modulation drive signal generates a positive current on phase A. This current is detected by the controller, and the S 11 No failure occurred, S 22 Failure occurred; turn off S 11 , generate a 5ms PWM signal to S 22 ; If the controller does not detect a positive current, it determines S 11 Normal, S 22 Fault; (2) Give S 11 A PWM drive signal does not generate a positive current on phase A, which means that S 11 If a fault occurs, continue to give S 22 The same PWM signal generates current on phase A at this time, which finally proves that S 11 Fault, S 22 normal; (3) Give S 11 A PWM drive signal, no forward current is detected on phase A, then it is determined that S 11 Failure, continue to give S 22 For the same PWM signal, if there is no current on phase A, then it is determined that S 11 and S 22 All failed.

8. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 6, characterized in that: In step 2, when the fault back electromotive force is positive, the PI structure is used to generate a PWM drive signal, and the drive signal is given to the switch tube S 11 ; Step 2.1.1: If no forward current is generated at this time, it can be determined that the switch tube S 11 Fault occurs, then switch S 11 The drive signal is blocked and the signal is given to the switch tube S 22 If a forward current is generated at this time, the switch tube S 22 No fault has occurred; if no current is generated, S 22 There was also a malfunction; Step 2.1.2: If a forward current is generated at this time, it can be determined that the switch tube S 11 If no fault occurs, the switch tube S 22 A fault may occur; turn on the switch S 11 The drive signal is blocked and the signal is given to the switch tube S 22 If no forward current is generated at this time, the switch tube S 22 A fault occurs; if a forward current is generated, then step 1 is judged wrong, and the switch tube S 11 and S 22 No failures occurred; Return to step 1 to continue judging; if current is generated twice, then the switch tube S 11 and S 22 All faults.

9. The method for diagnosing switch tube faults in a dual inverter system of an open-winding permanent magnet synchronous motor according to claim 6, characterized in that: In step 2, when the fault back electromotive force is negative, the PI structure is used to generate a PWM drive signal, and the drive signal is given to the switch tube S 12 ; Step 2.2.1: If no negative current is generated at this time, it can be determined that the switch tube S 12 Fault occurs, then switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If a negative current is generated at this time, the switch tube S 21 No fault has occurred; if no current is generated, S 21 A malfunction occurs; Step 2.2.2: If a negative current is generated at this time, it can be determined that the switch tube S 12 If no fault occurs, the switch tube S 21 A fault may occur; next, switch S 12 The drive signal is blocked and the signal is given to the switch tube S 21 If no negative current is generated at this time, the switch tube S 21 Fault occurs; if negative current is generated, then step 1 is judged wrong, and the switch tube S 12 and S 21 If no fault occurs, return to step 1 to continue the judgment.

Images