A method for diagnosing and locating single-phase ground faults in stator windings of large generators based on graded variable frequency pulse injection

Abstract

This invention discloses a method for diagnosing and locating single-phase grounding faults in the stator windings of large generators based on graded frequency-converter pulse injection. Addressing the problems of insufficient high-frequency signal strength leading to inaccurate fault severity detection and the inability of low-frequency signals to accurately pinpoint the fault location, this invention proposes a graded frequency-converter pulse injection fault location method. Low-frequency voltage pulses are used to determine the faulty phase and branch; then, high-frequency pulses are used to precisely determine the location of the single-phase grounding fault. This invention improves the speed and accuracy of fault location, effectively enhancing the maintenance efficiency of large generators and improving location precision.

Description

Technical Field

[0001] The invention relates to the field of relay protection design of power systems, and more particularly to a method for locating single-phase grounding faults of generator stator windings taking high fault resistance into consideration. Background Art

[0002] As the main power source in the system, large generators are key equipment in the power system, and the safety of their operation has always been highly valued. One of the main threats to the safe operation of generators comes from the insulation system. Accidents caused by insulation failures are often catastrophic and will cause huge losses. Studies have shown that accidents caused by insulation problems in stator windings account for about one-third of all accidents.

[0003] Since the deterioration process of stator insulation is irreversible and constantly accelerating, once partial insulation damage occurs, it is easy to further cause major faults such as single-phase grounding, turn-to-turn short circuit, and phase-to-phase short circuit. Among them, for single-phase grounding faults, conventional relay protection means cannot locate the fault. At the same time, the highly nonlinear, strong saturation and strong interference characteristics of hydro-turbine generators also increase the difficulty of single-phase grounding fault diagnosis and high-precision positioning. Therefore, a fault diagnosis method for high-precision positioning of the fault position has important engineering application value. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] A method for diagnosing and locating single-phase grounding faults of large generator stator windings based on graded variable frequency pulse injection, characterized in that it comprises the following steps:

[0006] S1, inject low-frequency voltage pulses into the phase winding, analyze the terminal phase current characteristics, and determine the fault phase;

[0007] S2, inject low-frequency voltage pulses into the phase winding, analyze the branch leakage current characteristics, and determine the fault branch;

[0008] S3, inject high-frequency voltage pulses, record the time when the traveling wave reaches the beginning and end respectively, and accurately locate the single-phase grounding fault.

[0009] Furthermore, in the step S1, a terminal current vector method is used to locate the phase fault, and a terminal diagnosis circuit and a pulse generation algorithm are designed.

[0010] The pulse generator is a single-phase full-bridge inverter circuit that can realize the forward and reverse injection of pulses, that is, the injected current pulse flows from the beginning of the winding to the end or from the end of the winding to the beginning. The pulse is generated by the switch tube S1 and the switch tube S2. The specific method is as follows: the drive signal of the switch tube S1 is generated by the current hysteresis controller, and the drive signal of the switch tube S2 is a periodic switch signal, which is used to control the period of the injected pulse, reduce the loss and torque generated by the current pulse injection, and prevent the rotor from rotating. After injecting the low-frequency voltage pulse, the current at the beginning of the winding (i h ) and terminal current (i e ), and calculate the difference between the ground current (i g ), and then comprehensively analyze i g and i e characteristics, determine the fault phase and fault degree. Under normal circumstances, i h Equal to i e ,i g Equal to 0; when a single-phase ground short circuit fault occurs, i h Greater than i e ,i g Not equal to 0.

[0011] The end current vector method makes use of the idea of ​​rotating vectors. The core of the construction is to convert the time coordinates of the end current into polar coordinates. When the motor is running, the current parameters change periodically with the operation of the winding. The current changes in this specific period are reflected in the polar coordinate diagram to construct a current vector diagram. There are overlapping areas between currents of different phases. The overlapping areas will not only reduce the intuitiveness of the current vector diagram, but also increase the difficulty of fault diagnosis. Therefore, it is considered to use the size contraction change to limit the current of each phase to 120°. In order to overcome the influence of different conduction intervals, the present invention defines a size contraction change coefficient k, which is used to compress the working interval of each phase to 120°. The selection of the size contraction change coefficient should take into account the most extreme case, that is, the maximum angle θ of the demagnetization interval corresponding to the moment when the maximum demagnetization interval occurs should be considered. max .

[0012] Furthermore, in step S2, a leakage current characteristic analysis method is used when monitoring the fault branch. When diagnosing, the voltage equation of the leakage current loop at time k-1 is:

[0013]

[0014] The voltage equation at time k is:

[0015]

[0016] Subtracting the two moment formulas yields:

[0017]

[0018] By comparing the calculation result of this formula with the normal inductance value, x can be obtained, thereby locating the fault.

[0019] Furthermore, in step S3, a high-frequency positioning method is used to accurately locate the single-phase grounding fault position, a high-frequency distributed parameter model is established, and a high-frequency voltage pulse is injected into the model to determine the fault position.

[0020] The time when the first high-frequency component is first reflected back to the head end is recorded as t1, the time when it first reaches the end is recorded as t2, and the total length of the fault branch is recorded as l. At this time, the calculation formula of x is:

[0021]

[0022] The extraction of t1 and t2 adopts the method based on the mutation of signal change rate. After the high-frequency positioning is completed, it is compared with the low-frequency signal to prevent diagnostic errors and ensure the effectiveness of diagnosis and positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a diagnostic flow chart of an embodiment of the present invention;

[0024] Figure 2 is a circuit diagram of a pulse generating device according to an embodiment of the present invention;

[0025] Figure 3 is a normalized curve diagram of the phase current excitation conduction time of an embodiment of the present invention;

[0026] Figure 4 is a schematic diagram of the calculation principle of the leakage current loop inductance according to an embodiment of the present invention;

[0027] Figure 5 This is a schematic diagram of the high-frequency pulse positioning principle of an embodiment of the present invention;

[0028] Figure 6 is a three-phase voltage waveform of the generator in the embodiment of the present invention when it is running at no load;

[0029] Figure 7 is a comparison of the air gap magnetic flux density of the generator during normal operation of the embodiment of the present invention and after a single-phase grounding fault occurs;

[0030] Figure 8 is the leakage current under different grounding resistances of the embodiment of the present invention;

[0031] Fig. 9 is a single-phase grounding equivalent model of the generator at low frequency according to an embodiment of the present invention;

[0032] Fig.10 is a high frequency distributed parameter model of an embodiment of the present invention;

[0033] Fig.11 This is the current distribution under 1Ω grounding resistance when the injection current is 1A and the bus voltage is 100V in the embodiment of the present invention;

[0034] Fig.12 This is the current distribution under 1Ω grounding resistance when the injection current is 10A and the bus voltage is 200V in the embodiment of the present invention;

[0035] Fig.13 This is a terminal current vector diagram of an embodiment of the present invention under a grounding resistance of 1Ω when an injection current of 1A and a bus voltage of 100V;

[0036] Fig.14 This is a terminal current vector diagram under a grounding resistance of 1Ω when the injection current is 10A and the bus voltage is 200V according to an embodiment of the present invention;

[0037] Fig.15 is the first-end current signal when the fault point of the embodiment of the present invention is 180 meters;

[0038] Fig.16 is the terminal current signal when the fault point of the embodiment of the present invention is 180 meters;

[0039] Fig.17 is the first-end current signal when the fault point is 300 meters in the embodiment of the present invention;

[0040] Fig.18 It is the terminal current signal when the fault point is 300 meters in the embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0042] Example

[0043] In order to more clearly illustrate the purpose, technical solutions and advantages of the present invention, the following will be described in detail with reference to the accompanying drawings. It should be noted that the specific embodiments mentioned are only used to explain the present invention and do not limit the scope of the present invention. In addition, in each embodiment of the present invention described below, as long as there is no conflict between the technical features, they can be combined with each other.

[0044] It should be emphasized that, according to the actual implementation requirements, the various steps / components described in this application can be split into more detailed steps / components, or two or more steps / components can be flexibly combined to form new steps / components.

[0045] This embodiment provides a method for diagnosing and locating a single-phase ground fault of a large generator stator winding based on graded variable frequency pulse injection, which is characterized by comprising the following steps:

[0046] S1, inject low-frequency voltage pulses into the phase winding, analyze the terminal phase current characteristics, and determine the fault phase;

[0047] S2, inject low-frequency voltage pulses into the phase winding, analyze the branch leakage current characteristics, and determine the fault branch;

[0048] S3, inject high-frequency voltage pulses, record the time when the traveling wave reaches the beginning and end respectively, and accurately locate the single-phase grounding fault.

[0049] Furthermore, in the step S1, a terminal current vector method is used to locate the phase fault, and a terminal diagnosis circuit and a pulse generation algorithm are designed.

[0050] The pulse generating device is a single-phase full-bridge inverter circuit, such as Figure 2 As shown. It can realize the positive and negative injection of pulses, that is, the injected current pulse flows from the beginning of the winding to the end or from the end of the winding to the beginning. The pulse is generated by the switch tube S1 and the switch tube S2. The specific method is as follows: the driving signal of the switch tube S1 is generated by the current hysteresis controller, and the driving signal of the switch tube S2 is a periodic switching signal, which is used to control the period of the injected pulse, reduce the loss and torque generated by the current pulse injection, and prevent the rotor from rotating. After injecting the low-frequency voltage pulse, the current at the beginning of the winding (i h ) and terminal current (i e ), and calculate the difference between the ground current (i g ), and then comprehensively analyze i g and i e characteristics, determine the fault phase and fault degree. Under normal circumstances, i h Equal to i e ,i g Equal to 0; when a single-phase ground short circuit fault occurs, i h Greater than i e ,i gis not equal to 0. After the fault occurs, as the grounding resistance changes, a single current feature cannot meet the requirements of high-precision diagnosis and positioning. First, the characteristics of ie are analyzed. Under normal circumstances, the end current of phase A, the end current of phase B and the end current of phase C are symmetrical. Once a single-phase grounding fault occurs, the symmetry of the current is destroyed. Therefore, the present invention proposes a diagnostic method based on the end current vector diagram.

[0051] The construction of the current vector diagram uses the idea of ​​rotating vectors. The core of the construction is to convert the time coordinates of the terminal current into polar coordinates. When the motor is running, the current parameters change periodically with the winding operation. This change is reflected in the polar coordinate diagram to construct the current vector diagram. There are overlapping areas between different phase currents. Therefore, it is considered to use the scale change to limit the current of each phase within 120°. Figure 3 As shown, the angle corresponding to the moment when the winding starts to be excited is θ on , the corresponding angle when the winding enters the demagnetization interval is θ off , conduction angle θ c The corresponding angle is θ c =θ off -θ on , the larger the conduction angle, the larger the corresponding working range. In order to overcome the influence of different conduction ranges, the present invention defines a size contraction coefficient k, which is used to compress the working range of each phase to within 120°. The selection of the size contraction coefficient should consider the most extreme case, that is, the maximum demagnetization interval angle θ corresponding to the moment when the maximum demagnetization interval occurs should be considered. max , the expression of the coefficient of change k is:

[0052]

[0053] Furthermore, in step S2, a leakage current characteristic analysis method is used when monitoring the fault branch. The calculation principle diagram is as follows: Figure 4 As shown, the voltage equation of the leakage current loop during diagnosis is:

[0054]

[0055] The voltage equation at time k is:

[0056]

[0057] Subtracting the two moment formulas yields:

[0058]

[0059] By comparing the calculation result of this formula with the normal inductance value, x can be obtained, thereby locating the fault.

[0060] Furthermore, in step S3, a high-frequency positioning method is used to accurately locate the single-phase grounding fault position, a high-frequency distributed parameter model is established, and a high-frequency voltage pulse is injected into the model to determine the fault position.

[0061] When high-frequency pulses are injected, the positioning principle diagram is as follows: Figure 5 As shown. The time when the first high-frequency component is first reflected back to the head end is recorded as t1, the time when it first reaches the end is recorded as t2, and the total length of the fault branch is recorded as l. At this time, the calculation formula of x is:

[0062]

[0063] The extraction of t1 and t2 adopts the method based on the mutation of signal change rate. After the high-frequency positioning is completed, it is compared with the low-frequency signal to prevent diagnostic errors and ensure the effectiveness of diagnosis and positioning.

[0064] Furthermore, to ensure the effectiveness of the present invention, simulation analysis was carried out, a finite element model of a large hydroelectric generator was built in Altair Flux software, and simulations of normal operation of the generator under no-load conditions and under ground fault were performed; a simulation model of a large salient-pole generator under normal and ground fault conditions was established in Matlab / Simulink.

[0065] like Figure 6 The three-phase voltage waveform when the generator is running at no load is shown; Figure 7 The figure shows the comparison of the air gap flux density of the generator during normal operation and after a single-phase ground fault occurs; Figure 8 The figure shows the leakage current under different grounding resistances.

[0066] like Fig. 9 The figure shows the equivalent model of single-phase grounding of the generator at low frequency; Fig.10 It is a distributed parameter model at high frequency.

[0067] Fig.11 The figure shows the current distribution under 1Ω grounding resistance when the injection current is 1A and the bus voltage is 100V; Fig.12 The current distribution under 1Ω grounding resistance when the injection current is 10A and the bus voltage is 200V; Fig.13 It is the terminal current vector diagram under 1Ω grounding resistance when the injection current is 1A and the bus voltage is 100V; Fig.14 It is the terminal current vector diagram under 1Ω grounding resistance when the injection current is 10A and the bus voltage is 200V.

[0068] Furthermore, the total length of a single branch line rod is set to 360 meters and a high-frequency pulse is injected at the head end. When the single-phase grounding fault point is at 180 meters, the head end current signal and the terminal end voltage signal results are as follows: Fig.15 ,16 As shown; when the single-phase grounding fault point is 300 meters, the first-end current signal and the terminal voltage signal results are as follows Fig.17 , 18 shown.

Claims

1. A method for diagnosing and locating single-phase grounding faults of large generator stator windings based on graded variable frequency pulse injection, characterized in that: The following steps are involved: Inject low-frequency voltage pulses into the phase winding, analyze the terminal phase current characteristics, and determine the fault phase; Inject low-frequency voltage pulses into the phase windings, analyze branch leakage current characteristics, and identify the faulty branch; Inject high-frequency voltage pulses and record the time when the traveling wave reaches the beginning and end respectively, so as to accurately locate the single-phase grounding fault.

2. According to claim 1, a method for diagnosing and locating single-phase ground fault of stator winding of large generator based on graded variable frequency pulse injection is characterized in that: In step S1, a terminal current vector method is used to locate the phase fault, and a terminal diagnosis circuit and a pulse generation algorithm are designed.

3. The method for diagnosing and locating single-phase ground fault of stator winding of a large generator based on graded variable frequency pulse injection according to claim 1 is characterized in that: In step S2, a leakage current characteristic analysis method is used when monitoring the fault branch.

4. The method for diagnosing and locating single-phase ground fault of stator winding of a large generator based on graded variable frequency pulse injection according to claim 1 is characterized in that: In step S3, a high-frequency positioning method is used to accurately locate the single-phase grounding fault position, a high-frequency distributed parameter model is established, and a high-frequency voltage pulse is injected into the model to determine the fault position.

5. The terminal diagnosis circuit and pulse generation algorithm of the terminal current vector method according to claim 2, characterized in that: The pulse generator is a single-phase full-bridge inverter circuit that can realize the forward and reverse injection of pulses. The pulse is generated by the switch tube S1 and the switch tube S2. The specific method is as follows: the drive signal of the switch tube S1 is generated by the current hysteresis controller, and the drive signal of the switch tube S2 is a periodic switch signal, which is used to control the cycle of the injected pulse, reduce the loss and torque generated by the current pulse injection, and prevent the rotor from rotating. After the low-frequency voltage pulse is injected, the head-end current (i h ) and terminal current (i e ), and calculate the difference between the ground current (i g ), and then comprehensively analyze i g and i e characteristics, determine the fault phase and fault degree.

6. A terminal current vector method according to claim 2, characterized in that: The terminal current vector method uses the idea of ​​rotating vectors. The core of the construction is to convert the time coordinates of the terminal current into polar coordinates. When the motor is running, the current parameters change periodically with the operation of the winding. The current changes in this specific period are reflected in the polar coordinate diagram to construct a current vector diagram. The overlapping areas between currents of different phases will increase the difficulty of fault diagnosis. Therefore, it is considered to use the size contraction change to limit the current of each phase to 120°. In order to overcome the influence of different conduction intervals, the present invention defines a size contraction change coefficient k, which is used to compress the working interval of each phase to 120°.

7. A leakage current characteristic analysis method according to claim 3, characterized in that: The voltage equation of the leakage current loop during diagnosis is as follows: The voltage equation at time k is: Subtracting the two moment formulas yields: By comparing the calculation result of this formula with the normal inductance value, x can be obtained, thereby locating the fault.

8. The high frequency positioning method according to claim 4, characterized in that: The time when the first high-frequency component is first reflected back to the head end is recorded as t1, the time when it first reaches the end is recorded as t2, and the total length of the fault branch is recorded as l. At this time, the calculation formula of x is The extraction of t1 and t2 adopts the method based on the mutation of signal change rate. After the high-frequency positioning is completed, it is compared with the low-frequency signal to prevent diagnostic errors and ensure the effectiveness of diagnosis and positioning.

Images