Fault diagnosis method for H-bridge half-controlled rectifier of doubly salient electro-magnetic generator

Abstract

The invention discloses a fault diagnosis method for an H-bridge half-controlled rectifier of an electro-magnetic doubly salient generator, and relates to the field of electro-magnetic doubly salient generators, the method is developed based on an improved H-bridge half-controlled rectifier topological structure, and the improved H-bridge half-controlled rectifier topological structure is additionally provided with a current sensor for each phase of armature winding. The leads of the left and right bridge arms connected with each phase of armature winding are wound on the current sensor in a winding manner, so that each phase of current sensor can obtain more current information; open-circuit fault diagnosis can be realized through current measured by a current sensor wound by a working two-phase armature winding in each sector, the method is small in calculation amount and easy to realize, is not influenced by rotating speed or load change, is also not influenced by fault occurrence positions and conduction angle change, can realize single-tube, double-tube and multi-tube fault positioning, and can be applied to the field of fault diagnosis. The diagnosis effect is good; and the application range is wide.

Description

Technical Field

[0001] The present application relates to the field of electrically excited double-pole generators, and in particular to a fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited double-pole generator. Background Art

[0002] The doubly salient electromagnetic generator (DSEG) has the advantages of simple structure, high reliability and strong adaptability to high temperature and high speed operating environment, and has significant advantages in voltage regulation and fault demagnetization. It has broad application prospects in the fields of aviation, wind power, etc. Compared with the traditional uncontrolled rectifier system, the DSEG controlled rectifier system based on the H-Bridge Semi-Controlled Rectifier (HBSCR) adjusts the armature current by changing the conduction angle of the switch tube. Because its three-phase current is uncoupled, the commutation overlap problem is light, which can significantly improve the motor output power, reduce losses and optimize motor performance. Therefore, it has been favored by many scholars.

[0003] However, affected by the harsh working environment, HBSCR is the weakest link in the entire power generation system, usually manifested as short-circuit faults and open-circuit faults. Open-circuit faults will lead to deterioration of DC voltage quality, reduced system load capacity, and reduced power generation. Short-circuit faults will cause serious overcurrent in the phase winding, causing serious damage to the system. It has the characteristics of fast response and great harm, so it is necessary to respond quickly to short-circuit faults. The current conventional method is to convert it into an open-circuit fault through hardware circuits. Therefore, in order to improve the reliability of the DSEG system, it is of great significance to study the diagnosis method of the open-circuit fault of the HBSCR switch tube. Summary of the invention

[0004] In view of the above problems and technical requirements, this application proposes a fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator. The technical solution of this application is as follows:

[0005] A fault diagnosis method for an H-bridge semi-controlled rectifier of an electrically excited double-pole generator, wherein the lower bridge arm wire of the left bridge arm connected to any M-phase armature winding in the electrically excited double-pole generator passes through an M-phase current sensor CSM from a P pole to an N pole and is wound one turn, the upper bridge arm wire of the left bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from an N pole to a P pole and is wound two turns, the upper bridge arm wire of the right bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from a P pole to an N pole and is wound two turns, and the lower bridge arm wire of the right bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from an N pole to a P pole and is wound one turn; M=A, B, C respectively represent the three-phase armature windings in the electrically excited double-pole generator, and the current positive direction is from the bridge arm midpoint of the right bridge arm connected to any M-phase armature winding to the bridge arm midpoint of the left bridge arm;

[0006] The fault diagnosis method of the H-bridge half-controlled rectifier of the electrically excited doubly salient generator includes:

[0007] Get the current i of the current sensor wound by the bridge arm connected to the forward working winding Z in the sector where the rotor position angle θ is currently located CSZ , and the current i of the current sensor wound by the bridge arm connected to the negative working winding Q in the current sector CSQ ; The positive working winding Z and the negative working winding Q are two-phase armature windings in the electrically excited double-pole generator;

[0008] The lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z Z1 The lower bridge arm switch tube T of the right bridge arm connected to the negative working winding Q Q2 During the conduction process, when the current i CSZ In case of abnormality, determine the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z. Z1 An open circuit fault occurs. When the current i CSQ In case of abnormality, determine the lower bridge arm switch tube T in the right bridge arm connected to the negative working winding Q Q2 An open circuit fault has occurred.

[0009] The beneficial technical effects of this application are:

[0010] The present application discloses a fault diagnosis method for an H-bridge semi-controlled rectifier of an electrically excited double-pole generator. The method is based on an improved H-bridge semi-controlled rectifier topology structure. The improved H-bridge semi-controlled rectifier topology structure adds a current sensor to each phase armature winding, and the wires of the left and right bridge arms connected to each phase armature winding are wound around the current sensor by winding, so that each phase current sensor can obtain more current information. The current provided by the current sensor wound by the working two-phase armature winding in each sector can realize open circuit fault diagnosis. The calculation amount is small and easy to implement. The method is not affected by the speed or load changes, nor is it affected by the fault location and the conduction angle changes. The applicability of the diagnosis method is broadened and it has the characteristics of strong robustness. The method can realize single-tube, double-tube and multi-tube fault location, and can be extended to the open circuit fault diagnosis of the switch tube in the N-phase DSEG controllable rectifier system, with a wide range of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is the topology of the traditional HBSCR-based electrically excited doubly salient generator controlled rectifier system.

[0012] Figure 2 This is the topology of the HBSCR-based electrically excited double-salient-pole generator controllable rectification system of the present application.

[0013] Figure 3 It is a control strategy diagram of the HBSCR-based electrically excited doubly salient-pole generator controlled rectifier system in one electrical angle cycle.

[0014] Figure 4 It is the current loop of the A-phase armature winding in the forward energy storage and current increasing stage in sector 1.

[0015] Figure 5 It is the current loop of the A-phase armature winding in the forward freewheeling power generation stage in sector 1.

[0016] Figure 6 In sector 1, when the armature winding of phase A is connected to the lower bridge arm switch tube T of the left bridge arm A1 Waveform of phase A current during normal operation.

[0017] Figure 7 This is the system current flow diagram before 0° commutation.

[0018] Figure 8 It is the current loop of the C-phase winding formed in sector 1 when the positive C-phase current has been continued to 0 at 0°.

[0019] Fig. 9 It is the current loop of the C-phase winding formed in sector 1 when there is a positive C-phase current at 0°.

[0020] Fig.10 In sector 1, [0°,θ c ] interval, the lower bridge arm switch tube T A1 Waveform diagram during normal operation, and the lower bridge arm switch tube T A1 Current loop and waveform diagram when an open circuit fault occurs.

[0021] Fig.11 It is the current loop of the negative freewheeling power generation stage of the C-phase armature winding in sector 1.

[0022] Fig.12 In sector 1, when the armature winding of phase C is connected to the lower bridge arm switch tube T of the right bridge arm C2 Waveform of phase C current during normal operation.

[0023] Fig.13 When the end point θ1 of the forward current commutation of phase C in sector 1 is 0°, the lower bridge arm switch tube T C2 Waveform diagram during normal operation, and the lower bridge arm switch tube T C2 Current loop and waveform diagram when an open circuit fault occurs.

[0024] Fig.14 When the forward current commutation end point of phase C in sector 1 is 0°<θ1<θ c In the case of C2 Waveform diagram during normal operation, and the lower bridge arm switch tube T C2 Current loop and waveform diagram when an open circuit fault occurs.

[0025] Fig.15 When the forward current commutation end point of phase C in sector 1 is θ1>θ c In the case of C2 Waveform diagram during normal operation.

[0026] Fig.16 The present invention is a flow chart of a method for diagnosing faults of an H-bridge half-controlled rectifier of an electrically excited doubly salient generator in an embodiment. DETAILED DESCRIPTION

[0027] The specific implementation of the present application is further described below in conjunction with the accompanying drawings.

[0028] The topology of the conventional H-bridge half-controlled rectifier HBSCR-based electrically excited doubly salient generator controlled rectifier system is as follows: Figure 1As shown, the electrically excited double-salient-pole generator includes three-phase armature windings, which are respectively recorded as A, B, and C phases. One end of any M-phase armature winding is connected to the midpoint of the bridge arm on the left side, and the other end is connected to the midpoint of the bridge arm on the right side. M = A, B, and C represent the three-phase armature windings in the electrically excited double-salient-pole generator, respectively. m Represents the phase current of the M-phase armature winding, m=a, b, c respectively corresponds to the case where the M-phase armature winding is the A, B, C phase armature winding, and the current is in the positive direction from the midpoint of the right bridge arm connected to any M-phase armature winding to the midpoint of the left bridge arm. The structure of each bridge arm is the same, a bridge arm includes an upper bridge arm diode and a lower bridge arm switch tube, and both ends of the lower bridge arm switch tube are also anti-parallel diodes, the anode of the upper bridge arm diode is connected to the collector of the lower bridge arm switch tube and is used as the midpoint of the bridge arm to connect the armature winding, the cathode of the upper bridge arm diode is connected to the output positive electrode, the emitter of the lower bridge arm switch tube is connected to the output negative electrode, the load R is connected between the output positive electrode and the output negative electrode and provides the output voltage u o , a filter capacitor C is also connected between the output positive electrode and the output negative electrode. For the convenience of distinction, the upper bridge arm diode in the left bridge arm connected to any M-phase armature winding is recorded as D M1 , the lower bridge arm switch is denoted as T M1 , the lower bridge arm switch tube T M1 The anti-parallel diode is denoted as D M2 The upper bridge arm diode in the right bridge arm connected to the M-phase armature winding is denoted as D M3 , the lower bridge arm switch is denoted as T M2 , the lower bridge arm switch tube T M2 The anti-parallel diode is denoted as D M4 .

[0029] This application optimizes this traditional topological structure, and the optimized topological structure adopted is as follows Figure 2 As shown, the topological structure used in the present application adds one-phase current sensor to each of the three-phase armature windings, and the current sensor added to any M-phase armature winding is recorded as CSM. The lower bridge arm wire of the left bridge arm connected to any M-phase armature winding passes through the M-phase current sensor CSM from the P pole to the N pole and is wound one turn, the upper bridge arm wire of the left bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from the N pole to the P pole and is wound two turns, the upper bridge arm wire of the right bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from the P pole to the N pole and is wound two turns, and the lower bridge arm wire of the right bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from the N pole to the P pole and is wound one turn. The upper bridge arm wire of each bridge arm refers to the wire between the midpoint of the bridge arm and the upper bridge arm switch tube, and the lower bridge arm wire of each bridge arm refers to the wire between the midpoint of the bridge arm and the lower bridge arm switch tube.

[0030] This topology can enable each phase current sensor to obtain more current information by winding. Any M-phase current sensor CSM can obtain the lower bridge arm switch tube T in the left bridge arm connected to the M-phase armature winding. M1 The current i flowing through TM1 , the upper bridge arm diode D in the left bridge arm connected to the M-phase armature winding M1 The current i flowing through DM1 , the upper bridge arm diode D in the right bridge arm connected to the M-phase armature winding M3 The current i flowing through DM3 , the lower bridge arm switch tube T in the right bridge arm connected to the M-phase armature winding M2 The current i flowing through TM2 ,like Figure 2 As shown. The M-phase current sensor takes the positive direction of current from the P pole to the N pole, so the currents of the three current sensors are written as:

[0031]

[0032] That is, the current i of the current sensor CSM wound by the bridge arm connected to any M-phase armature winding CSM Written as i CSM =i TM1 -2i DM1 +2i DM3 -i TM2 .

[0033] After adjusting the topology structure, the conduction state of each lower bridge arm switch tube in the H-bridge half-controlled rectifier of the controllable rectifier system is as follows: Figure 3 As shown, the controlled rectifier system of the electrically excited double-salient-pole generator takes 0° to 360° as a control cycle and is divided into sector 1 of 0° to 120°, sector 2 of 120° to 240°, and sector 3 of 240° to 360°. Figure 3 In the figure, m = a, b, c represent the armature windings of phases A, B, and C respectively, and L m is the self-inductance of the M-phase armature winding, L mf is the mutual inductance between the M-phase armature winding and the field winding, e mf is the excitation back EMF of the M-phase armature winding, θ is the rotor position angle, θ c is the conduction angle of the switch tube in the sector, Figure 3 When the driving signal of any lower bridge arm switch is high, it means that the lower bridge arm switch is turned on, otherwise the lower bridge arm switch is turned off. Since the three sectors are symmetrical to each other, all the following analyses take the working process in sector 1 of 0° to 120° as an example, and the analysis of the other two sectors can be easily obtained by analogy.

[0034] 1. Working process of phase A winding in sector 1 of 0°~120°

[0035] 1. The lower bridge arm switch tube T of the left bridge arm connected to the A-phase armature winding A1 Under normal operation

[0036] In sector 1, e af >0, in order to make the current direction consistent with the polarity of the induced potential, it is necessary to turn on the lower bridge arm switch tube T of the left bridge arm connected to the A-phase armature winding. A1 To increase the current change rate, increase the armature current, and convert it into magnetic field energy for storage. Therefore, in [0°,θ c ] interval, the lower bridge arm switch tube T A1 In the on state, the current loop is: D A4 →A phase winding→T A1 →D A4 ,like Figure 4 At this time, the A-phase armature winding is in the forward energy storage and current rising stage, and it can be obtained from formula (1):

[0037] i CSA =i TA1 -i TA2 =i a -(-i a )=2i a >0 (2)

[0038] In [θ c ,120°] interval, the lower bridge arm switch tube T A1 Turn off, in order to continue the original positive A phase current of the armature winding, the phase current passes through the diode D A4 and D A1 A freewheeling loop is formed, and the current loop is: load R→D A4 →A phase winding→D A1 → Load R, such as Figure 5 As shown. According to the current loop, the following equation can be written:

[0039] e a -i a r=u o (3)

[0040] In the formula, e a is the induced potential of the armature winding of phase A, r is the internal resistance of the armature winding, and the current change rate corresponding to formula (3) is:

[0041]

[0042] In the formula, e ar is the A-phase magnetic resistance back EMF. Considering i a r compared to e af 、e ar and uo is small, so it can be ignored within the error range. Depends on af 、e ar and u o The relationship between the three is due to u o >0, therefore, e af +e ar -u o The amplitude can be positive or negative, making In [θ c ,120°] interval has three situations:

[0043] (a) In this case, the phase current i of the A-phase armature winding is a like Figure 6 As shown in (a), i a ≠0, i CSA =-2i DA1 -i TA2 =-2i a -(-i a )=-i a <0.

[0044] (b) and In this case, the phase current i of the armature winding of phase A is smaller. a like Figure 6 As shown in (b), i a ≠0, i CSA =-2i DA1 -i TA2 =-2i a -(-i a )=-i a <0.

[0045] (c) and In this case, the phase current i of the armature winding of phase A is larger. a like Figure 6 As shown in (c) in this case [θ c ,120°] interval will appear i a After it drops to 0 and remains at 0, this is because when i a When it just drops to 0, e ar and i a r is 0, and substituting it into formula (4) shows that e af o , that is, the excitation back EMF is less than the output voltage. Since phase A cannot provide energy to the load, i a Keep it at 0. ​

[0046] Combining the three situations, we can know that in [θ c ,120°], when the A-phase armature winding is in the forward freewheeling power generation stage, i CSA ≤0.

[0047] In summary, in sector 1 from 0° to 120°, in [0°,θ c ] interval, when the A-phase armature winding is in the forward energy storage and current rising stage, i CSA =2i a >0. In [θ c ,120°], when the A-phase armature winding is in the forward freewheeling power generation stage, i CSA =-i a ≤0.

[0048] 2. The above analysis does not take into account the commutation overlap of the system. The following is an analysis of the commutation process near 0°. To simplify the analysis, the current change rate in this part ignores the smaller winding internal resistance. Before 0° commutation, the system current flows as follows: Figure 7 As shown, if commutation overlap is considered, further analysis should be performed on this basis.

[0049] Under the APC strategy, T A1 Opening will pass through loop D A4 →A phase winding→T A1 →D A4 Generate positive phase A current. If the positive phase C current has been flowing to 0 before 0°, then T C2 Will be like Figure 8 As shown by loop D C2 →C phase winding→T C2 →D C2 Generate negative C-phase current. If there is still positive C-phase current at 0°, then Fig. 9 As shown by diode D C1 and D C4 Commutation, T C2 Will be D C2 Reverse clamped, in zero voltage turn-on state.

[0050] From the above analysis, it can be seen that in sector 1, the current of phase A is not affected by the commutation overlap, and the current waveform is still as Figure 6 shown.

[0051] 3. Analysis of T A1 The impact of the failure.

[0052] T A1 During normal operation, [0°,θ c ]In the interval T A1 Drive signal, i a and iCSA The current waveform is as follows Fig.10 As shown in (a) in .

[0053] When T A1 In [0°,θ c ] When a fault occurs in the section, the current loop will be A4 →A phase winding→T A1 →D A4 ” is converted into “load R→D A4 →A phase winding→D A1 →Load R”, the working state of phase A is transformed from the forward energy storage and current rising state to the forward freewheeling power generation state, so that i CSA From positive to negative.

[0054] When T A1 In [θ c ,360°] section when a fault occurs, T A1 It does not work in this range and has no fault characteristics. However, in the next electrical angle cycle [0°,θ c ] interval, the voltage at the A phase terminal u a The expression is as follows:

[0055]

[0056] Among them, e aa is the self-inductance induced potential of phase A, and its value is

[0057] When θ=0°, i a is 0, and substituting it into formula (5) yields u a =e af According to e af with u o There are two possibilities for the size relationship:

[0058] When e af >u o When , phase A generates electricity in an uncontrolled rectifier mode. Fig.10 As shown in (b), i a >0,i CSA Current waveform Fig.10 As shown in (c) in .

[0059] When e af o When the A phase winding cannot provide energy to the load, Fig.10 As shown in (d) in the figure, i a Keep it at 0, i CSA The waveform is shown in (e) in FIG10 .

[0060] Therefore, it can be determined that when T​A1 When a fault occurs, the c ] interval detected i CSA ≤0.

[0061] 2. Working process of C phase winding in sector 1 of 0°~120°

[0062] In sector 1, e cf <0, similar to the forward energy storage current rising stage, in order to make the current direction consistent with the polarity of the induced potential, it is necessary to turn on the lower bridge arm switch tube T of the right bridge arm connected to the C-phase armature winding. C2 To increase the current change rate, increase the armature current, and convert it into magnetic field energy for storage. Therefore, in [0°,θ c ] interval, the lower bridge arm switch tube T C2 In the on state, the current loop is: D C2 →C phase winding→T C2 →D C2 ,like Figure 8 As shown, at this time, the C-phase armature winding is in the negative energy storage and current rising stage. By the same logic as formula (1), we can get i CSC =i TC1 -i TC2 =-(-i c )-(-i c )=2i c <0.

[0063] In [θ c ,120°] interval, the lower bridge arm switch tube T C2 In order to continue the original positive phase C current of the armature winding, the phase current passes through the diode D C3 and D C2 A freewheeling loop is formed, and the current loop is: load R→D C2 →C phase winding→D C3 → Load R, such as Fig.11 As shown. At this time, the C-phase armature winding is in the negative freewheeling power generation stage, and the following equation can be written according to the current loop:

[0064] e c -i c r=-u o (6)

[0065] In the formula, e c is the induced potential of the C-phase armature winding, r is the internal resistance of the armature winding, and the current change rate corresponding to equation (6) is shown as follows:

[0066]

[0067] In the formula, e cr is the C phase magnetoresistance back EMF. Considering ic r compared to e cf 、e cr and u o is small, so it can be ignored within the error range. Depends on cf 、e cr and u o The relationship between the three is due to

[0068] u o >0, therefore, e cf +e cr -u o The amplitude can be positive or negative, making In [θ c ,120°] interval has three situations:

[0069] (a) In this case, the phase current i of the C-phase armature winding is c like Fig.12 As shown in (a), i c ≠0, i CSC =i TC1 +2i DC3 =-(-i c )+2(-i c )=-i c >0.

[0070] (b) and The phase current i of the C-phase armature winding in this case is smaller. c like Fig.12 As shown in (b), i c ≠0, i CSC =i TC1 +2i DC3 =-(-i c )+2(-i c )=-i c >0.

[0071] (c) and In this case, the phase current i of the C-phase armature winding is larger. c like Fig.12 As shown in (c) in this case [θ c ,120°] interval will appear i c After rising to 0 and remaining at 0, this is because when i c When it just rises to 0, e cr and i c r is 0, and substituting it into formula (6) shows that -ecf o , that is, the amplitude of the excitation back EMF is less than u o , so phase C cannot provide energy to the load, i c Keep it at 0.

[0072] From the above three situations, we can know that in [θ c ,120°], when the C-phase armature winding is in the negative freewheeling generation stage, i CSC ≥0.

[0073] 2. Similar to the A-phase winding, the above analysis does not take into account the system's commutation overlap. Figure 7 to Figure 9 Partial analysis shows that when considering commutation overlap, the current waveform of phase C is related to the end point θ1 of the forward current commutation of phase C in sector 1, and the conduction angle θ c The relationship with θ1 will also have an impact, so in sector 1 [0°,θ c ] interval i c There are three current situations:

[0074] (1) When θ1 = 0°, T C2 Normally, [0°,θ c ]T in the interval C2 Drive signal, i c and i CSC The current waveform is as follows Fig.13 As shown in (a) in .

[0075] (2) When 0°<θ1<θ c When T C2 Normally, [0°,θ c ]T in the interval C2 Drive signal, i c and i CSC The current waveform is as follows Fig.14 As shown in (a) in .

[0076] (3) When θ c When <θ1<120°, T C2 Normally, [0°,θ c ]T in the interval C2 Drive signal, i c and i CSC The current waveform is as follows Fig.15 shown.

[0077] 3. The following analyzes T in the above three cases one by one C2 The impact of an open circuit fault.

[0078] (1) In Fig.13 In the case of θ1 = 0° shown​

[0079] When T C2 In [0°,θ c ] When a fault occurs in the section, the current loop will be C2 →C phase winding→T C2 →D C2 ” becomes “Load R→D C2 →C phase winding→D C3 →Load R”, the working state of phase C changes from negative energy storage to negative continuous current generation, i CSC From negative to positive.

[0080] When T C2 In [θ c ,360°] when a fault occurs, due to the interval T C2 does not work, so there is no fault characteristic. But at T C2 The next electrical angle cycle after the fault occurs [0°,θ c ] interval, the voltage at the C phase terminal u c The expression is as follows:

[0081]

[0082] In the formula, e cc is the self-inductance induced potential of phase C, and its value is When θ=0°, i c is 0, and substituting it into formula (8) yields u c =e cf According to e cf with u o There are two possibilities for the size relationship:

[0083] when|e cf |>u o When , phase C generates electricity in an uncontrolled rectifier mode, such as Fig.13 As shown in (b), i c <0.

[0084] when|e cf | o When the C phase winding cannot provide energy to the load, Fig.13 As shown in (c), i c Keep it at 0.

[0085] Therefore, in this case, when T C2 When a fault occurs, at [0°,θ c ] The fault feature i can be extracted from the interval CSC ≥0.

[0086] (2) In Fig.14 0°<θ1<θ shown​c In the case of

[0087] When T C2 When a fault occurs in the interval [0°,θ1], considering that the current of phase C is in phase switching, the current loop is “load R→D C4 →C phase winding→D C1 → Load R", such as Fig.14 As shown in (b) in the figure. CSC =-i c <0, we can find that T C2 D C4 Reverse clamp, so T C2 The fault does not affect the current loop in this section. When θ=θ1, i c =0, from formula (8) we can get u c =e cf According to e cf with u o There are two possibilities for the relationship between the size of |e cf |>u o When the C phase generates electricity in an uncontrolled rectifier mode, i c <0, then Fig.14 As shown in (c), at this time i CSC >0. when|e cf | o When the C phase winding cannot provide energy to the load, i c Keep it at 0, then Fig.14 As shown in (d) in the figure, at this time i CSC = 0. Therefore, based on the above two possibilities, there is a fault feature i CSC ≥0.

[0088] When T C2 In [θ1,θ c ] If a fault occurs in the interval, the current loop will be C2 →C phase winding→T C2 →D C2 ” becomes “Load R→D C2 →C phase winding→D C3 →Load R", such as Fig.14 As shown in (e), the working state of phase C changes from negative energy storage to negative continuous current generation. CSC From negative to positive, such as Fig.14 As shown in (f) in the figure. At this time, there is also a fault feature i CSC ≥0.

[0089] When T C2 In [θ c ,360°] When a fault occurs, due to T C2 ​It does not work in this range, so it will be in the next electrical angle cycle [0°,θ c ] and make the diagnosis according to the above method.

[0090] In summary, in this case, when T C2 When a fault occurs, at [0°,θ c ] The fault feature i can be extracted from the interval CSC ≥0.

[0091] (3) In Fig.15 The θ shown c In the case of <θ1<120°

[0092] When T C2 In [0°,θ c ] When a fault occurs in the section, Fig.15 As shown in [0°,θ c ]Interval i c ≥0, we know that T C2 is clamped, so similar to the above analysis, this interval cannot detect the occurrence of faults.

[0093] When T C2 In [θ c ,360°] When a fault occurs, due to T C2 It does not work in this range, so it will be in the next electrical angle cycle [0°,θ c ] for diagnosis, and from the above, we can see that this situation is in [0°,θ c ]In the interval T C2 will be clamped, making T C2 Failures will not be affected by the system.

[0094] Therefore, in this case, T C2 The fault has no impact on the system and can be ignored.

[0095] Combining the above three situations, we can know that when 0°≤θ1<θ c In the case of T C2 Faults will have an impact on the system and can be in [0°,θ c ] The fault feature i is extracted within the interval CSC ≥0.

[0096] Based on the above, T A1 and T C2 The fault analysis can be summarized as follows:

[0097] When T A1 During normal operation, the armature winding of phase A is in [0°,θ c ] is in the positive energy storage and flow increase stage, i CSA =2ia >0. Phase A armature winding [θ c ,120°] interval is in the forward continuous current generation stage, i CSA =-i a ≤0. When T A1 When an open circuit fault occurs, it will be in [0°,θ c ] interval detected i CSA ≤0.

[0098] When T C2 In normal operation, the armature winding of phase C is in [0°,θ c ] is in the negative energy storage and flow increase stage, i CSC =2i c <0. The armature winding of phase C is [θ c ,120°] interval is in the negative follow-up power generation stage, i CSC =-i c ≥0. When T C2 When an open circuit fault occurs, it will be in [0°,θ c ] interval detected i CSC ≥0.

[0099] For sector 2, T B1 and T A2 The fault analysis of T in sector 3 is similar. C1 and T B2 The fault analysis of the three sectors is similar, so the fault analysis of the three sectors can be summarized as follows:

[0100] Any sector has its corresponding positive working winding Z and negative working winding Q. In the electrical angle interval [θ s ,θ e ], the positive working winding Z is connected to the lower bridge arm switch tube T in the left bridge arm Z1 And the lower bridge arm switch tube T in the right bridge arm connected to the negative working winding Q Q2 Will be in [θ s ,θ s +θ c ] interval, and in [θ s +θ c ,θ e ] interval. The positive working winding Z and the negative working winding Q are two-phase armature windings in the electrically excited double-pole generator:

[0101] For sector 1, the electrical angle range it covers is 0° to 120°. The positive working winding Z of this sector is the A-phase winding, the negative working winding Q is the C-phase winding, and T Z1 =T A1 , T Q2 =TC2 .

[0102] For sector 2, the electrical angle range it covers is 120° to 240°. The positive working winding Z is the B-phase winding, the negative working winding Q is the A-phase winding, and T Z1 =T B1 , T Q2 =T A2 .

[0103] For sector 3, the electrical angle range it covers is 240° to 360°. The positive working winding Z is the C-phase winding, the negative working winding Q is the B-phase winding, and T Z1 =T C1 , T Q2 =T B2 .

[0104] Through the above analysis, it can be determined that the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z in each sector Z1 The lower bridge arm switch tube T of the right bridge arm connected to the negative working winding Q Q2 At the same time, during the conduction process, the current i of the current sensor wound by the bridge arm connected to the forward working winding Z is obtained. CSZ It can detect the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z. Z1 Whether an open circuit fault occurs, and the current i of the current sensor wound by the bridge arm connected to the negative working winding Q obtained CSQ It can detect the lower bridge arm switch tube T in the right bridge arm connected to the negative working winding Q. Q2 Whether an open circuit fault occurs, specifically:

[0105] When the lower bridge arm switch tube T of the current sector Z1 And the lower bridge arm switch tube T Q2 At the same time, during the conduction interval, the current i of the current sensor wound by the bridge arm connected to the forward working winding Z is detected. CSZ >0, the lower bridge arm switch tube T can be determined Z1 Normal operation. When the current i CSZ When ≤0, the current i can be determined CSZ Abnormal, thus determining the lower bridge arm switch tube T Z1 An open circuit fault has occurred.

[0106] When the lower bridge arm switch tube T of the current sector Z1 And the lower bridge arm switch tube T Q2 At the same time, during the conduction interval, the current i of the current sensor wound by the bridge arm connected to the negative working winding Q is detected. CSQ <0, determine the lower bridge arm switch tube T Q2Normal operation. When the current i CSQ When ≥0, the current i can be determined CSQ Abnormal, confirm the lower bridge arm switch tube T Q2 An open circuit fault has occurred.

[0107] However, considering that the current sensor has measurement errors in practice, so that it cannot sample zero current, in one embodiment, the current i is not directly CSZ and current i CSQ Compared with zero current, a positive parameter ε is introduced as the current threshold. In one embodiment, i th <ε no ,i th is the maximum value of the current sensor measurement error when the phase current is 0, i no It is the current amplitude of the electrically excited doubly salient pole generator running at no load.

[0108] Then, when the lower bridge arm switch tube in the left bridge arm connected to the positive working winding Z and the lower bridge arm switch tube in the right bridge arm connected to the negative working winding Q are turned on at the same time:

[0109] When the lower bridge arm switch tube T of the current sector Z1 And the lower bridge arm switch tube T Q2 At the same time, the current i is detected in the conduction interval CSZ <ε to determine the current i CSZ Abnormal, thus determining that the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z in the current sector Z1 An open circuit fault has occurred.

[0110] When the lower bridge arm switch tube T of the current sector Z1 And the lower bridge arm switch tube T Q2 At the same time, the current i is detected in the conduction interval CSQ >-ε when the current i is determined CSQ Abnormal, thus determining the lower bridge arm switch tube T in the right bridge arm connected to the negative working winding Q in the current sector Q2 An open circuit fault has occurred.

[0111] For further information, please refer to Fig.10 As shown in (a) in sector 1, when T A1 Normal working time a The current waveform is given by Fig.10 It can be seen that in T A1 During normal operation, CSA The current waveform starts from 0 and gradually increases. Assuming i CSA The time taken to rise from 0 to ε is t1, and -ε will appear during this period. CSA <ε, this feature is similar to T​​A1 When an open circuit fault occurs, CSA <ε partially overlap, that is, in T A1 Normal work may result in misjudgment. Fig.14 0°<θ1<θ shown c The t2 period in the case and Fig.13 In the case of θ1 = 0°, misjudgment also occurs during the time period t3.

[0112] In order to solve the above misjudgment problem, the lower bridge arm switch tube T is introduced. Z1 The open circuit fault detection time threshold δ1 and the lower bridge arm switch tube T Q2 The open-circuit fault detection time threshold δ2 is used, and the optimized fault diagnosis strategy is to switch the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z. Z1 The lower bridge arm switch tube T of the right bridge arm connected to the negative working winding Q Q2 During the conduction process:

[0113] When the current i CSZ <-ε, or, the detected current -ε≤i CSZ When the duration t of <ε reaches the time threshold δ1, the current i is determined CSZ Abnormal, thus determining that the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z in the current sector Z1 An open circuit fault has occurred.

[0114] When the current i CSQ >ε, or, current -ε is detected CSQ When the duration t of ≤ε reaches the time threshold δ2, the current i is determined CSQ Abnormal, thus determining the lower bridge arm switch tube T in the right bridge arm connected to the negative working winding Q in the current sector Q2 An open circuit fault has occurred.

[0115] From this, we can define the lower bridge arm switch tube T Z1 Fault flag F Z1 And the lower bridge arm switch tube T Q2 Fault flag F Q2 for:

[0116]

[0117] Moreover, since the three phases of the DSEG controlled rectifier system based on HBSCR operate independently, the multi-tube fault can be regarded as the superposition of the single-tube open-circuit fault, that is, the multi-tube fault characteristics are the superposition of the single-tube open-circuit fault characteristics. Z1 And the lower bridge arm switch tube T Q2 ​The diagnostic results can be superimposed, and the final diagnostic result can be: the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z Z1 An open circuit fault occurs, or the negative working winding Q is connected to the lower bridge arm switch tube T of the right bridge arm Q2 An open circuit fault occurs and the switch is in a single-tube fault state. Alternatively, the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z Z1 An open circuit fault occurs, and the negative working winding Q is connected to the lower bridge arm switch tube T of the right bridge arm Q2 An open circuit fault occurs and the double-tube fault state occurs. The diagnostic logic is as follows: Fig.16 shown.

[0118] Next, the lower bridge arm switch tube T Z1 The open circuit fault detection time threshold δ1 and the lower bridge arm switch tube T Q2 The method for determining the value of the open circuit fault detection time threshold δ2 is as follows:

[0119] (1) Lower bridge arm switch tube T Z1 The open circuit fault detection time threshold δ1

[0120] As can be seen from the above analysis, in sector 1, it is mainly in i CSA The process of the current waveform gradually rising from 0 to ε will generate a current to the lower bridge arm switch tube T A1 Therefore, for any sector, in order to avoid misdiagnosis, it is sufficient to ensure that the time threshold δ1>t1, where t1 is the time when the H-bridge half-controlled rectifier is in normal operation [θ s ,θ s +θ c ]Interval current i CSZ The time taken to rise from 0 to ε.

[0121] The following methods for calculating the time t1 include:

[0122] Taking sector 1 as an example, the current loop of phase A in the time period t1 is "D A4 →A phase winding→T A1 →D A4 According to Kirchhoff's voltage law (KVL), we can get i a The rate of change is as follows:

[0123]

[0124] Considering that i a Very small, so and i a The amplitude of r is very small and can be ignored within the error range. In addition, due to the excitation self-inductance L f is large and t1 is short, so if Considered unchanged, It can be taken as 0 within the error range. Finally, formula (11) is simplified to:

[0125]

[0126] Since t1 is very short, equation (12) can be further discretized as follows:

[0127]

[0128] The same is true for other sectors. Therefore, for the forward working winding Z in any sector, the phase current i of the forward working winding Z can be determined according to Kirchhoff's voltage law. z The change rate of the H-bridge half-controlled rectifier under normal operating conditions is discretized and obtained as follows:

[0129]

[0130] Among them, z=a, b, c respectively corresponds to the case where the forward working winding Z is the A, B, C phase armature winding. L zf is the mutual inductance between the forward working winding Z and the excitation winding, L z is the self-inductance of the forward working winding Z, i f is the excitation current and ω is the rotor electrical angular velocity.

[0131] Then Δt=t1、Δi z =Δi CSZ / 2=ε / 2Substituting into formula (14) we can obtain:

[0132]

[0133] (2) Lower bridge arm switch tube T Q2 The open circuit fault detection time threshold δ2

[0134] As can be seen from the above analysis, in sector 1, Fig.13 and Fig.14 The situation will cause the lower bridge arm switch tube T C2 Therefore, the value of the time threshold δ2 needs to take these two situations into consideration to ensure that both misdiagnosis situations are excluded. Therefore, the time threshold δ2>max(t2,t3). Where:

[0135] (a) t2 is the end point of the positive current commutation of the negative working winding Q. θ1 is located at T Z1 and T Q2 In the case of the conduction process, when the H-bridge half-controlled rectifier is in normal operation [θ s ,θ s +θc ]Interval current i CSQ The total time taken to rise from -ε to 0 at the end point of the forward current commutation θ1 and then fall back to -ε. c It is the conduction angle of the lower bridge arm switch of the right bridge arm connected to the negative working winding Q in the current sector.

[0136] Combination Fig.14 The situation is analyzed by taking the C phase winding in sector 1 as an example. Fig.14 It can be seen that t2 = t 21 +t 22 ,in:

[0137] t 21 The current loop of phase C during the time period is: load R→D C4 →C phase winding→D C1 →Load R, according to Kirchhoff's voltage law (KVL), we can get i c The rate of change is as follows:

[0138]

[0139] Similar to the above analysis of phase A, within the error range, equation (16) can be simplified to:

[0140]

[0141] Further discretization of (17) yields:

[0142]

[0143] The same is true for other sectors. Therefore, for the negative working winding Q in any sector, the phase current i of the negative working winding Q can be determined according to Kirchhoff's voltage law. q The rate of change before the end point θ1 of the forward current commutation is discretized to obtain:

[0144]

[0145] Among them, q=a, b, c corresponds to the case where the negative working winding Q is the A, B, C phase armature winding respectively. L qf is the mutual inductance between the negative working winding Q and the excitation winding, L q is the self-inductance of the negative working winding Q, i f is the excitation current, ω is the rotor electrical angular velocity, u o is the output voltage.

[0146] Then Δt=t 21 , Δi q =-Δi CSQ=-εSubstituting into formula (19) we can get:

[0147]

[0148] Similar to the above analysis, according to Kirchhoff's voltage law, the phase current i of the negative working winding Q is determined under the normal operating state of the H-bridge half-controlled rectifier. q The rate of change after the forward current commutation end point θ1 is discretized to obtain:

[0149]

[0150] Δt=t 22 , Δi q =-Δi CSQ =-ε Substituting into equation (21) we can get the current i CSQ The time taken for the forward current to decrease from 0 at the end point θ1 to -ε is:

[0151]

[0152] Finally, the time taken can be determined

[0153] (b) t3 is when the positive current commutation end point θ1 of the negative working winding Q is at the beginning of the sector, and the H-bridge half-controlled rectifier is in normal operation [θ s ,θ s +θ c ]Interval current i CSQ The time taken to decrease from 0 to -ε.

[0154] contrast Fig.13 and Fig.14 It can be seen that the current loop in the t3 period is consistent with the above t 22 The same time period, so the phase current i of the negative working winding Q is determined according to Kirchhoff's voltage law q The change rate under the normal operation of the H-bridge half-controlled rectifier is discretized to obtain the form shown in the above formula (21), and then Δt=t3, Δi q =-Δi CSQ =-εSubstituting into the equation, we get:

[0155]

[0156] It can be seen that t3 is different from the above t 22 are the same.

[0157] Based on the above calculations, considering that t1, t2 and t3 are very short and all near 0°, we can set:

[0158]

[0159] Among them, L min is the minimum value of the three-phase armature winding self-inductance, L max It is the maximum value of the three-phase armature winding self-inductance.

[0160] The above is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.

Claims

1. A fault diagnosis method for an H-bridge semi-controlled rectifier of an electrically excited doubly salient generator, characterized in that: The lower bridge arm wire of the left bridge arm connected to any M-phase armature winding in the electrically excited double-pole generator passes through the M-phase current sensor CSM from the P pole to the N pole and is wound one turn, the upper bridge arm wire of the left bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from the N pole to the P pole and is wound two turns, the upper bridge arm wire of the right bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from the P pole to the N pole and is wound two turns, and the lower bridge arm wire of the right bridge arm connected to the M-phase armature winding passes through the M-phase current sensor CSM from the N pole to the P pole and is wound one turn; M=A, B, C respectively represent the three-phase armature windings in the electrically excited double-pole generator, and the current positive direction is from the bridge arm midpoint of the right bridge arm connected to any M-phase armature winding to the bridge arm midpoint of the left bridge arm; The fault diagnosis method of the H-bridge half-controlled rectifier of the electrically excited doubly salient generator comprises: Get the current i of the current sensor wound by the bridge arm connected to the forward working winding Z in the sector where the rotor position angle θ is currently located CSZ , and the current i of the current sensor wound by the bridge arm connected to the negative working winding Q in the current sector CSQ ; The positive working winding Z and the negative working winding Q are two-phase armature windings in the electrically excited double-pole generator; The lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z Z1 The lower bridge arm switch tube T of the right bridge arm connected to the negative working winding Q Q2 During the conduction process, when the current i CSZ In case of abnormality, determine the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z. Z1 An open circuit fault occurs. When the current i CSQ In case of abnormality, determine the lower bridge arm switch tube T in the right bridge arm connected to the negative working winding Q Q2 An open circuit fault has occurred.

2. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 1, characterized in that: The M-phase current sensor has the P pole pointing to the N pole as the positive direction of the current. The current i of the current sensor wound by the bridge arm connected to any M-phase armature winding CSM =i TM1 -2i DM1 +2i DM3 -i TM2 , where i TM1 Indicates the lower bridge arm switch tube T in the left bridge arm connected to the M-phase armature winding M1 The current flowing through, i DM1 Indicates the upper bridge arm diode D in the left bridge arm connected to the M-phase armature winding M1 The current flowing through, i DM3 Indicates the upper bridge arm diode D in the right bridge arm connected to the M-phase armature winding M3 The current flowing through, i TM2 Indicates the lower bridge arm switch tube T in the right bridge arm connected to the M-phase armature winding M2 The current flowing through Detection current i CSZ and current i CSQ Whether the abnormality includes: When the lower bridge arm switch tube T of the current sector Z1 And the lower bridge arm switch tube T Q2 At the same time, the current i is detected in the conduction interval CSZ <ε to determine the current i CSZ Abnormal, current i detected CSQ >-ε when the current i is determined CSQ Exception; where ε is a positive parameter.

3. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 2, characterized in that: Detection current i CSZ and current i CSQ Whether the abnormality also includes: When the lower bridge arm switch tube T of the current sector Z1 And the lower bridge arm switch tube T Q2 At the same time, the current i is detected in the conduction interval CSZ <-ε, or, the current detected is -ε≤i CSZ When the duration of <ε reaches the time threshold δ1, the current i is determined CSZ abnormal; When the lower bridge arm switch tube T of the current sector Z1 And the lower bridge arm switch tube T Q2 At the same time, the current i is detected in the conduction interval CSQ >ε, or, current -ε is detected CSQ When the duration of ≤ε reaches the time threshold δ2, the current i is determined CSQ abnormal.​ 4. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 3, characterized in that: The time threshold δ1>t1, t1 is the current i under normal operation of the H-bridge half-controlled rectifier CSZ The time taken to rise from 0 to ε.

5. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 4, characterized in that: Methods for determining the time t1 include: According to Kirchhoff's voltage law, the phase current i of the forward working winding Z is determined z The change rate of the H-bridge half-controlled rectifier under normal operating conditions is obtained by discretization. Let Δt = t1, Δi z =Δi CSZ / 2=ε / 2Substituting this into in, z=a, b, c respectively represent the case where the forward working winding Z is the armature winding of phase A, B, C, L zf is the mutual inductance between the forward working winding Z and the excitation winding, L z is the self-inductance of the forward working winding Z, i f is the excitation current and ω is the rotor electrical angular velocity.

6. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 3, characterized in that: Time threshold δ2>max(t2,t3), where: t2 is the end point of the positive current commutation of the negative working winding Q. θ1 is located at T Z1 and T Q2 In the case of the conduction process, the H-bridge half-controlled rectifier is in normal operation when the current i CSQ The total time taken to rise from -ε to 0 at the end point of the forward current commutation θ1 and then fall back to -ε; where θ c It is the lower bridge arm switch tube T of the right bridge arm connected to the negative working winding Q Q2 The conduction angle of the switch in the current sector; t3 is the current i in the normal operation state of the H-bridge half-controlled rectifier when the positive current commutation end point θ1 of the negative working winding Q is at the beginning of the sector. CSQ The time taken to decrease from 0 to -ε.

7. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 6, characterized in that: Methods for determining the time t2 include: According to Kirchhoff's voltage law, the phase current i of the negative working winding Q is determined under the normal operating state of the H-bridge half-controlled rectifier. q The rate of change before the end point θ1 of the forward current commutation is discretized to obtain Δt=t 21 , Δi q =-Δi CSQ =-εSubstitute into the current i CSQ The time taken to rise from -ε to 0 at the end point of the forward current commutation θ1 According to Kirchhoff's voltage law, the phase current i of the negative working winding Q is determined under the normal operating state of the H-bridge half-controlled rectifier. q The rate of change after the forward current commutation end point θ1 is discretized to obtain Δt=t 22 , Δi q =-Δi CSQ =-εSubstitute into the current i CSQ The time taken for the forward current to decrease from 0 at the end point θ1 to -ε Determine the time Among them, q=a, b, c respectively represent the negative working winding Q is the A, B, C phase armature winding, L qf is the mutual inductance between the negative working winding Q and the excitation winding, L q is the self-inductance of the negative working winding Q, i f is the excitation current, ω is the rotor electrical angular velocity, u o is the output voltage.

8. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 6, characterized in that: Methods for determining the time t3 include: According to Kirchhoff's voltage law, the phase current i of the negative working winding Q is determined q The change rate of the H-bridge half-controlled rectifier under normal operating conditions is obtained by discretization. Let Δt = t3, Δi q =-Δi CSQ =-εSubstituting into Among them, q=a, b, c respectively represent the negative working winding Q is the A, B, C phase armature winding, L qf is the mutual inductance between the negative working winding Q and the excitation winding, L q is the self-inductance of the negative working winding Q, i f is the excitation current and ω is the rotor electrical angular velocity.

9. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 2, characterized in that: i th <ε no ,i th is the maximum value of the current sensor measurement error when the phase current is 0, i no It is the current amplitude of the electrically excited doubly salient pole generator running at no load.​ 10. The fault diagnosis method for an H-bridge half-controlled rectifier of an electrically excited doubly salient generator according to claim 1, characterized in that: In the sector where the rotor position angle θ is currently located: The forward working winding Z is connected to the lower bridge arm switch tube T in the left bridge arm Z1 An open circuit fault occurs, or The negative working winding Q is connected to the lower bridge arm switch tube T of the right bridge arm. Q2 An open circuit fault occurs and the device is in a single-tube fault state; Alternatively, the lower bridge arm switch tube T in the left bridge arm connected to the forward working winding Z Z1 An open circuit fault occurs, and the negative working winding Q is connected to the lower bridge arm switch tube T of the right bridge arm Q2 An open circuit fault occurs and the device is in a double-tube fault state.

Images