A method and device for off-line diagnosis of rotor winding short-circuit fault of variable speed pumped storage unit
By injecting pulse signals into the rotor winding of variable speed pumping accumulator and collecting voltage feedback, combining characteristic waveforms and VMD-TEO methods, accurately identifying and positioning the rotor winding short circuit faults, the problem of diagnosis in the prior art is solved, and maintenance efficiency is improved. Security.
Patent Information
- Application Number
- CN202210909621.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-29
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2042-07-29
AI Technical Summary
The prior art cannot accurately diagnose the fault type and fault location of the rotor winding short circuit fault of the variable speed pumping accumulator unit, resulting in slow maintenance speed and affecting the safe and stable operation of the unit.
By injecting three-phase low-voltage pulse signals into the end of the rotor winding, the three-phase voltage feedback signal is collected, the characteristic waveform is constructed, and the fault type and positioning fault location are determined by combining the maximum peak time difference detection method and the VMD-TEO method.
Accurate identification and positioning of short circuit faults of the rotor winding of variable speed pumping accumulator unit, covering about 90% of the rotor winding length, and the positioning error does not exceed the single-turn length, significantly speeding up the maintenance speed and ensuring the safe and stable operation of the unit.
Smart Images

Figure CN115291138B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of generator fault diagnosis, and more specifically, relates to an offline diagnosis method and device for a variable speed pumped storage unit rotor winding short circuit fault. Background Art
[0002] With the large-scale grid connection of renewable energy sources such as wind power, the power system urgently needs pumped storage power stations to play a role in rapid response such as peak shaving and valley filling, frequency and phase modulation to ensure the safe and reliable operation of the power grid. Compared with traditional fixed-speed pumped storage units, pumped storage units based on variable-speed doubly-fed induction motors have a higher unit operating range and operating efficiency, and can effectively suppress low-frequency oscillations in the system and maintain grid stability. Variable-speed pumped storage technology can provide better support for the decarbonization of the power system, and the construction of variable-speed pumped storage power stations has become a new trend.
[0003] The fixed-speed pumped storage unit uses a synchronous motor with DC excitation, and the rotor winding short-circuit fault form is only a turn-to-turn short circuit. The variable-speed pumped storage unit is essentially a doubly fed induction motor with three-phase AC excitation, but it is different from the traditional doubly fed induction motor. Its rotor winding adopts a multi-branch structure. Therefore, the form and fault characteristics of its rotor winding short-circuit fault are more complicated. Although the excitation structures of the two units are different, the rotor winding short circuit will be reflected in the short-circuit point overheating, the winding temperature rise, the rotor excitation current increase, the air gap magnetic flux imbalance causing abnormal vibration of the unit, and even the accident of burning the rotor winding and the body. The frequent rotor winding short-circuit faults pose a great threat to the safety of the unit itself.
[0004] In terms of rotor winding fault detection of variable-speed pumped storage units, existing methods mostly focus on detecting the electrical quantity fed back to the stator side when the rotor winding short-circuit fault occurs. They can only detect the fault but cannot further obtain fault information such as the fault type and fault phase. They cannot obtain the exact location of the fault, which is not conducive to speeding up fault inspection and repair. Summary of the invention
[0005] In view of the defects in the prior art, the present invention provides an offline diagnosis method for rotor winding short-circuit fault of a variable speed pumped storage unit, which aims to diagnose the rotor winding short-circuit fault of the variable speed pumped storage unit and provide accurate fault information for maintenance personnel to speed up the fault repair and provide guarantee for the safe and stable operation of the variable speed pumped storage unit.
[0006] To achieve the above object, according to one aspect of the present invention, a variable speed pumped storage unit rotor winding short circuit fault offline diagnosis method is provided. After the variable speed pumped storage unit stops after the rotor winding short circuit fault occurs, the fault offline diagnosis method comprises the following steps:
[0007] S1. The pulse generator injects the same low-voltage pulse signal into the three-phase winding through the three-phase collector ring of the rotor winding, and collects the three-phase feedback voltage signal at the end of the rotor winding, which is recorded as u A 、u B 、u C , make a difference between the three-phase voltage signals and get three characteristic waveforms u AB 、u BC 、u CA ;
[0008] S2. Determine whether the three characteristic waveforms are activated, and use the number of activated characteristic waveforms to determine whether the rotor winding short-circuit fault type is a phase-to-phase short circuit or a turn-to-turn short circuit; if it is a turn-to-turn short circuit fault, proceed to step S3; if it is a phase-to-phase short circuit fault, proceed to step S4;
[0009] S3. First, the fault phase of the turn-to-turn short circuit is determined by the characteristics of the activation characteristic waveform, and then the fault phase is selected as the characteristic waveform formed by the subtracted phase, and then the maximum peak time difference detection method is further used to determine whether the turn-to-turn short circuit fault type is the same branch turn-to-turn short circuit or the same phase different branch turn-to-turn short circuit, and then enter step S5;
[0010] S4. Using the characteristics of the activated characteristic waveform to determine the faulty two phases when a phase-to-phase short circuit occurs;
[0011] S5. Select the corresponding fault phase voltage acquisition signal u according to the fault type and fault phase x And characteristic waveform u xy , using VMD-TEO method to locate the voltage signal u x And characteristic waveform u xy The moment corresponding to the initial mutation position is used to determine the fault location through the time difference between the two and the measured wave velocity.
[0012] Furthermore, the step S2 specifically includes:
[0013] S201. Determine whether the maximum absolute value of the three characteristic waveforms is greater than the set value (100mV). If it is greater than the set value, it is defined as the characteristic waveform is activated, otherwise it is defined as the characteristic waveform is not activated;
[0014] S202. There are only two situations in which the number of activated characteristic waveforms exists. If all three characteristic waveforms are activated, the fault type is determined to be a phase-to-phase short circuit. If only two characteristic waveforms are activated, the fault type is determined to be a turn-to-turn short circuit.
[0015] Furthermore, the step S3 specifically includes:
[0016] S301. When the fault type is a turn-to-turn short circuit fault, the phases contained in the two simultaneously activated characteristic waveforms are identified as fault phases;
[0017] S302. Select the fault phase as the characteristic waveform formed by the subtracted phase, that is, corresponding to phases A, B, and C, respectively select characteristic waveforms u AB ,u BC and u CA ;
[0018] S303. Use the maximum peak time difference detection method to determine the type of inter-turn short circuit fault. The maximum peak time difference detection method is:
[0019]
[0020] Among them, T is the propagation time of the traveling wave on the length of the two branches of the rotor winding, which serves as the threshold; t0 is the injection pulse positioning time; u is the selected characteristic waveform; t(max(u)) is the time corresponding to the positive maximum value on the selected characteristic waveform.
[0021] Furthermore, the step S4 specifically includes:
[0022] When the fault type is a phase-to-phase short circuit fault, the two phases corresponding to the largest amplitudes among the three characteristic waveforms are identified as the two faulty phases.
[0023] Furthermore, in step S5, if the fault type is a short circuit between turns of the same branch or a short circuit between turns of the same phase but different branches, the fault phase voltage signal is selected (for example, if the A phase is faulty, u is selected). A ) and one of the characteristic waveforms of the fault phase (e.g., if phase A fails, u AB or CA ) to carry out the next step of fault location identification; if the fault type is interphase short circuit, then randomly select one of the two fault phase voltage signals (such as AB phase fault, u A or B ) and the characteristic waveform of two fault phases (such as AB phase fault, select u AB ) to proceed to the next step of fault location identification.
[0024] Furthermore, the step S5 specifically includes:
[0025] S501. Using VMD-TEO method to locate the selected fault phase voltage signal u x The time corresponding to the starting mutation position is the injection pulse time;
[0026] S502. Using VMD-TEO method to locate the selected characteristic waveform u xy The time corresponding to the starting mutation position is the arrival time of the reflected traveling wave at the fault point;
[0027] S503. Determine the distance between the fault point and the rotor winding head end by using the difference between the injection pulse time and the arrival time of the reflected traveling wave at the fault point and the measured wave velocity, which is
[0028]
[0029] Where v is the measured wave velocity, and Δt is the time difference between the injection pulse and the arrival time of the reflected traveling wave at the fault point.
[0030] The main steps for locating VMD-TEO faults are as follows:
[0031] (1) Collect the fault voltage traveling wave signal or characteristic signal at the head-end detection point;
[0032] (2) Perform VMD decomposition on the signal to obtain K IMF components;
[0033] (3) Select the high-frequency modal components in the IMF components for TEO analysis to obtain the Teager instantaneous energy spectrum;
[0034] (4) The moment corresponding to the first mutation extreme point of the instantaneous energy spectrum is the arrival moment of the traveling wave.
[0035] The measured wave velocity can be obtained by the following method:
[0036] The low-voltage pulse signal is injected into only one phase through the collector ring of the rotor winding, and the pulse signal is received in any of the other two phases. The difference between the injection signal and the reception signal time is the propagation time of the traveling wave on the length of the two branches of the rotor winding. The traveling wave speed can be obtained as
[0037]
[0038] Wherein, L is the length of a single branch of the rotor winding, t0 is the injection pulse positioning time, and t1 is the receiving pulse positioning time of the other two phases. t0 and t1 can be obtained by the VMD-TEO method.
[0039] Another aspect of the present invention provides an offline diagnostic device for rotor winding short-circuit fault of a variable speed pumped storage unit, comprising:
[0040] The pulse generation module is used to generate three low-voltage pulse signals at the same time. The pulse signals are generated using the Heidler function model, and the generation of waveforms with different steepness can be achieved by adjusting the wave head time constant τ1 and the wave tail time constant τ2;
[0041] Impedance matching module, used to suppress the reflection of traveling waves at the end of the rotor winding to interfere with the fault signal characteristics, and the impedance matching effect can be achieved by adjusting the sliding rheostat;
[0042] Signal acquisition module, used to collect and display three-way voltage signals, and can store corresponding collected data;
[0043] The data analysis and calculation module obtains the data collected by the signal acquisition module and then processes and analyzes the collected data to implement the above method.
[0044] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:
[0045] (1) The present invention provides an offline diagnosis method and device for a short-circuit fault in the rotor winding of a variable-speed pumped storage unit. After a variable-speed pumped storage unit shuts down due to a short-circuit fault in the rotor winding, the same low-voltage pulse signal is injected into the rotor winding through the three-phase collector ring at the end of the rotor winding, and the three-phase end voltage feedback signal is collected, and then three characteristic waveforms are constructed. The characteristics of the three characteristic waveforms are used to comprehensively judge the fault type and fault phase. At the same time, for two types of inter-turn short-circuit faults that are difficult to distinguish, the maximum peak time difference detection method is used to identify the two fault types. This method can accurately identify the three fault types and fault phases of the short-circuit fault of the rotor winding of the variable-speed pumped storage unit, and the identification method can cover about 90% of the length of the rotor winding and is not affected by the fault resistance.
[0046] (2) The present invention uses a method combining variational mode decomposition and Teager energy (VMD-TEO) to detect the initial mutation time of the fault phase voltage signal and the characteristic waveform, and determines the location of the fault point based on the difference between the two and the measured wave velocity. This method can locate various types of short-circuit faults in the rotor winding of a variable-speed pumped storage unit, and the positioning error does not exceed the single-turn length of the rotor winding, which provides a strong guarantee for the rapid maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 A schematic flow chart of an offline diagnosis method for a variable speed pumped storage unit rotor winding short circuit fault provided by an embodiment of the present invention;
[0048] Figure 2 A schematic diagram of a rotor winding short-circuit fault model of a variable speed pumped storage unit provided in an embodiment of the present invention;
[0049] Figure 3 Three characteristic waveform diagrams of the rotor winding of the variable speed pumped storage unit under different short-circuit fault types provided by the embodiment of the present invention; (a1) is the same branch turn-to-turn short circuit u AB , (a2) is the short circuit between turns of the same branch u BC , (a3) is the short circuit between turns of the same branch u CA , (b1) is the short circuit between the same phase and different branches u AB , (b2) is the short circuit between the same phase and different branches u BC , (b3) is the short circuit between the same phase and different branches u CA , (c1) is the phase-to-phase short circuit u AB, (c2) is the phase-to-phase short circuit u BC , (c3) is the phase-to-phase short circuit u CA ;
[0050] Figure 4 An analysis diagram of the maximum peak generation time of a characteristic waveform provided by an embodiment of the present invention;
[0051] Figure 5 A flow chart for identifying the rotor winding short-circuit fault type and fault phase of a variable speed pumped storage unit provided by an embodiment of the present invention;
[0052] Figure 6 The positioning result diagram of the rotor winding of the variable speed pumped storage unit under different fault types provided in the embodiment of the present invention; (a) is a short circuit between the turns of the same branch, (b) is a short circuit between the turns of the same phase and different branches, and (c) is a phase-to-phase short circuit. DETAILED DESCRIPTION
[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0054] This embodiment provides an offline diagnosis method for a variable speed pumped storage unit rotor winding short circuit fault. The variable speed pumped storage unit shuts down after a rotor winding short circuit fault occurs. Figure 1 As shown, the fault offline diagnosis method includes the following steps:
[0055] S1. The pulse generator injects the same low-voltage pulse signal into the three-phase winding through the three-phase collector ring of the rotor winding, and collects the three-phase feedback voltage signal at the end of the rotor winding, which is recorded as u A 、u B 、u C ; Make a difference between the three-phase voltage signals and get three characteristic waveforms u AB 、u BC 、u CA .
[0056] S2. Determine whether the three characteristic waveforms are activated respectively, and use the number of activated characteristic waveforms to determine whether the rotor winding short-circuit fault type is a phase-to-phase short circuit or a turn-to-turn short circuit; if it is a turn-to-turn short circuit fault, go to step S3, if it is a phase-to-phase short circuit fault, go to step S4.
[0057] Specifically, the step S2 specifically includes:
[0058] S201. Determine whether the maximum absolute value of the three characteristic waveforms is greater than the set value (100mV). If it is greater than the set value, it is defined as the characteristic waveform is activated, otherwise it is defined as the characteristic waveform is not activated;
[0059] S202. If all three characteristic waveforms are activated, the fault type is determined to be a phase-to-phase short circuit; if only two characteristic waveforms are activated, the fault type is determined to be a turn-to-turn short circuit.
[0060] S3. First, use the characteristics of the activated characteristic waveform to determine the fault phase where the inter-turn short circuit occurs, then select the fault phase as the characteristic waveform composed of the subtracted phase, and then further use the maximum peak time difference detection method to determine whether the inter-turn short circuit fault type is the same branch inter-turn short circuit or the same phase different branch inter-turn short circuit.
[0061] Specifically, the step S3 specifically includes:
[0062] S301. When the fault type is a turn-to-turn short circuit fault, the phases contained in the two simultaneously activated characteristic waveforms are identified as fault phases;
[0063] S302. Select the fault phase as the characteristic waveform formed by the subtracted phase, that is, corresponding to phases A, B, and C, respectively select characteristic waveforms u AB ,u BC and u CA ;
[0064] S303. Use the maximum peak time difference detection method to determine the type of inter-turn short circuit fault. The maximum peak time difference detection method is:
[0065]
[0066] Among them, T is the propagation time of the traveling wave on the length of the two branches of the rotor winding, which serves as the threshold; t0 is the injection pulse positioning time; u is the selected characteristic waveform; t(max(u)) is the time corresponding to the positive maximum value on the selected characteristic waveform.
[0067] S4. Use the characteristics of the activated characteristic waveform to determine the two faulty phases when a phase-to-phase short circuit occurs.
[0068] Specifically, the step S4 specifically includes:
[0069] When the fault type is a phase-to-phase short circuit fault, the two phases corresponding to the largest amplitudes among the three characteristic waveforms are identified as the two faulty phases.
[0070] S5. Select the corresponding fault phase voltage acquisition signal u according to the fault type and fault phase x And characteristic waveform u xy , using VMD-TEO method to locate the voltage signal u x And characteristic waveform u xy The moment corresponding to the initial mutation position is used to determine the fault location through the time difference between the two and the measured wave velocity.
[0071] Specifically, in step S5, if the fault type is a short circuit between turns of the same branch or a short circuit between turns of the same phase and different branches, the fault phase voltage signal is selected (for example, if the A phase is faulty, u is selected). A ) and one of the characteristic waveforms of the fault phase (e.g., if phase A fails, u AB or CA ) to carry out the next step of fault location identification; if the fault type is interphase short circuit, then randomly select one of the two fault phase voltage signals (such as AB phase fault, u A or B ) and the characteristic waveform of two fault phases (such as AB phase fault, select u AB ) Carry out the next step of fault location identification;
[0072] Specifically, the step S5 includes:
[0073] S501. Using VMD-TEO method to locate the selected fault phase voltage signal u x The time corresponding to the starting mutation position is the injection pulse time;
[0074] S502. Using VMD-TEO method to locate the selected characteristic waveform u xy The time corresponding to the starting mutation position is the arrival time of the reflected traveling wave at the fault point;
[0075] S503. Determine the distance between the fault point and the rotor winding head end by using the difference between the injection pulse time and the arrival time of the reflected traveling wave at the fault point and the measured wave velocity, which is
[0076]
[0077] Where v is the measured wave velocity, and Δt is the time difference between the injection pulse and the arrival time of the reflected traveling wave at the fault point.
[0078] The measured wave velocity can be obtained by the following method:
[0079] The low-voltage pulse signal is injected into only one phase through the collector ring of the rotor winding, and the pulse signal is received in any of the other two phases. The difference between the injection signal and the reception signal time is the propagation time of the traveling wave on the length of the two branches of the rotor winding. The traveling wave speed can be obtained as
[0080]
[0081] Wherein, L is the length of a single branch of the rotor winding, t0 is the injection pulse positioning time, and t1 is the receiving pulse positioning time of the other two phases. t0 and t1 can be obtained by the VMD-TEO method.
[0082] Below, Figure 2Taking the variable speed pumped storage unit rotor winding short-circuit fault model shown as an example, the technical details of the present invention are explained. The model is established based on the structure of a variable speed unit actually used in a pumped storage power station. The rotor winding of the variable speed unit has three phases, each phase has two branches in parallel, each branch has 49 coils in series, each coil is 10.5m long, and the winding form is a double-layer wave winding. It should be understood that the variable speed pumped storage unit rotor winding short-circuit fault offline diagnosis method of the present invention is suitable for complex variable speed pumped storage units, and is not limited to Figure 2 The structures shown and the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0083] The variable speed pumped storage unit shuts down after a rotor winding short circuit fault occurs.
[0084] (1) The pulse generator injects the same low-voltage pulse signal into the three-phase winding through the three-phase collector ring of the rotor winding at the same time, and collects the three-phase feedback voltage signal at the end of the rotor winding, which is recorded as u A 、u B 、u C , and make a difference between the three-phase voltage signals, and get three characteristic waveforms u AB 、u BC 、u CA .
[0085] Among them, the signal injection position and the feedback voltage signal collection position are as follows: Figure 2 As shown, the signal acquisition module collects the voltage signal and stores it, and transmits the collected signal to the data analysis and calculation module, which performs a pairwise difference on the three-phase voltage signals to obtain three characteristic waveforms u AB 、u BC 、u CA .
[0086] (2) Determine whether the maximum absolute value of the three characteristic waveforms is greater than the set value (100mV). If it is greater than the set value, it is defined as the activation of the characteristic waveform; if only two characteristic waveforms are activated, the fault type is determined to be a turn-to-turn short circuit, and the process goes to step (3). If all three characteristic waveforms are activated, the fault type is determined to be a phase-to-phase short circuit, and the process goes to step (4).
[0087] In the embodiment, the three characteristic waveforms shown when the rotor winding of the variable speed pumped storage unit has three types of faults are as follows: Figure 3As shown. It can be obtained that, when a short circuit occurs between turns of the same branch, only two characteristic waveforms are activated among the characteristic waveforms a1, a2 and a3, so the fault type is judged to be a short circuit between turns, and the next step is to enter step (3); when a short circuit occurs between turns of the same phase and different branches, only two characteristic waveforms are activated among the characteristic waveforms b1, b2 and b3, so the fault type is judged to be a short circuit between turns, and the next step is to enter step (3); when a short circuit occurs between phases, all three characteristic waveforms are activated among the characteristic waveforms c1, c2 and c3, so the fault type is judged to be a short circuit between phases, and the next step is to enter step (4).
[0088] (3) First, the fault phase where the turn-to-turn short circuit occurs is determined by using the characteristics of the activated characteristic waveform, and then the fault phase is selected as the characteristic waveform formed by the subtracted phase, and then the turn-to-turn short circuit fault type is further determined to be the same-branch turn-to-turn short circuit or the same-phase different-branch turn-to-turn short circuit, and then the process proceeds to step (5);
[0089] When the fault type is a turn-to-turn short-circuit fault, the phases contained in the two simultaneously activated characteristic waveforms are respectively determined as fault phases; in the embodiment, when a turn-to-turn short-circuit fault occurs in the same branch of the rotor winding, the activated characteristic waveform is as follows: Figure 3 (a1) and Figure 3 As shown in (a3), the two activation characteristic waveforms u AB 、u CA The phases they both contain are phase A, so it is determined that the phase A has a short circuit between turns. When a short circuit occurs between the same phase and different branches of the rotor winding, the characteristic waveform is activated as follows: Figure 3 (b1) and Figure 3 As shown in (b3) in the figure, the activation characteristic waveform u AB 、u CA The commonly included phase is phase A, so it is determined that the inter-turn short circuit of phase A occurs.
[0090] Since only the fault phase is selected as the characteristic waveform formed by the subtracted phase, that is, corresponding to phases A, B, and C, the characteristic waveforms u AB ,u BC and u CA , and in the embodiment, both turn-to-turn short-circuit faults occur in phase A, so the characteristic waveform u AB is selected.
[0091] For two different turn-to-turn short-circuit faults, the maximum peak generation time of the selected characteristic waveform is analyzed as follows: Figure 4As shown. When there is a short circuit between turns of the same branch, the appearance of the maximum peak of the characteristic waveform is due to the reduction of the original waveform of the non-fault phase (reference phase) by the negative polarity reflected wave generated by the non-fault phase injection pulse at the fault point, and its propagation path is greater than the length of the two branches. The appearance of the maximum peak of the characteristic waveform of the short circuit fault between turns of the same phase and different branches is due to the increase of the original waveform of the fault phase by the positive polarity refracted wave generated by the fault phase injection pulse at the short circuit point, and its propagation path is less than the length of the two branches. Therefore, the maximum peak time difference detection method can be expressed as:
[0092]
[0093] Wherein, T is the propagation time of the traveling wave on the length of the two branches of the rotor winding, which is used as the threshold; t0 is the injection pulse positioning time; u is the selected characteristic waveform; t(max(u)) is the time corresponding to the positive maximum value on the selected characteristic waveform;
[0094] In the embodiment, the results of identifying two types of turn-to-turn short-circuit faults using the maximum peak time difference detection method are shown in Table 1. It can be seen that the maximum peak time difference detection method can accurately determine the type of turn-to-turn short-circuit fault under different fault distances and fault resistances. It should be noted that when the fault location is 450m away from the head end, the propagation time of the two branches is 5.03us, and the maximum peak time difference is 4.89us, which basically reaches the maximum length of identification. Therefore, this method can cover the range of about 90% of the length of the rotor winding and has high reliability.
[0095] Table 1
[0096]
[0097] (4) Using the activated waveform features to determine the two faulty phases with interphase short circuit; if the fault type is an interphase short circuit fault, the characteristic waveform amplitude obtained by subtracting the voltage signals collected from the two faulty phases is approximately twice the amplitude of the other two characteristic waveforms. Therefore, the two phases corresponding to the largest amplitude of the three characteristic waveforms are determined to be the two faulty phases;
[0098] In the embodiment, when a phase-to-phase short circuit fault occurs in the rotor winding, the characteristic waveform is activated as follows: Figure 3 (c1) Figure 3 (c2) and Figure 3 As shown in (c3), the activation characteristic waveform u AB The amplitude is the largest, so it is judged that the AB phases are short-circuited.
[0099] The fault type and fault phase identification process described in (2) to (4) is as follows: Figure 5 shown.
[0100] (5) Select the corresponding fault phase voltage acquisition signal u according to the fault type and fault phase xAnd characteristic waveform u xy If the fault type is a short circuit between turns of the same branch or a short circuit between turns of the same phase and different branches, select the fault phase voltage signal (for example, if the A phase is faulty, select u A ) and one of the characteristic waveforms of the fault phase (e.g., if phase A fails, u AB or CA ) to carry out the next step of fault location identification; if the fault type is interphase short circuit, then randomly select one of the two fault phase voltage signals (such as AB phase fault, u A or B ) and the characteristic waveform of two fault phases (such as AB phase fault, select u AB ) to proceed to the next step of fault location identification.
[0101] In the embodiment, for the actual A same branch turn-to-turn short circuit and A different branch turn-to-turn short circuit, the fault phase voltage signal u is selected A , characteristic waveform selection u AB The next step is to locate the fault. For the actual AB phase short circuit fault, select the fault phase voltage signal u A , characteristic waveform u AB Go to the next step of fault location and identification.
[0102] (6) Using VMD-TEO method to locate voltage signal u x And characteristic waveform u xy The moment corresponding to the initial mutation position is used to determine the fault location through the time difference between the two and the measured wave velocity.
[0103] The low-voltage pulse signal is injected into only one phase through the collector ring of the rotor winding, and the pulse signal is received in any of the other two phases. The difference between the injection signal and the reception signal time is the propagation time of the traveling wave on the length of the two branches of the rotor winding. The traveling wave speed can be obtained as
[0104]
[0105] Wherein, L is the length of a single branch of the rotor winding, t0 is the injection pulse positioning time, and t1 is the receiving pulse positioning time of the other two phases. t0 and t1 can be obtained by the VMD-TEO method.
[0106] In the embodiment, the low voltage pulse signal is injected only into phase A through the collector ring of the rotor winding, and the pulse signal is received in phase B, and the voltage signal u is located using the VMD-TEO method. A and u B The time corresponding to the starting mutation position is 0.28us and 5.31us respectively. The length of a single branch of the rotor winding is 514.5m, and the wave speed is 204.573m / us.
[0107] Using VMD-TEO method to locate the selected fault phase voltage signal u x The time corresponding to the starting mutation position is the injection pulse time; the VMD-TEO method is used to locate the selected characteristic waveform u xy The time corresponding to the starting mutation position is the arrival time of the reflected traveling wave at the fault point. Then, the distance between the fault point and the rotor winding head end is determined by using the difference time between the injection pulse time and the arrival time of the reflected traveling wave at the fault point and the measured wave speed, which is
[0108]
[0109] Where v is the measured wave velocity, and Δt is the time difference between the injection pulse and the arrival time of the reflected traveling wave at the fault point.
[0110] In the embodiment, the VMD-TEO method is used to locate the selected fault phase voltage signal u A And characteristic waveform u AB The moment corresponding to the starting mutation position of the pulse injection point and the arrival time of the reflected traveling wave at the fault point are obtained respectively. Then, according to the above formula, the fault location results of the three fault types at different fault distances and fault resistances are obtained as follows: Figure 6 As shown in the figure, the maximum absolute positioning errors of the same-phase branch turn-to-turn short-circuit fault, the same-phase branch turn-to-turn short-circuit fault and the phase-to-phase short-circuit fault are 3.79m, 8.66m and 4.03m, all of which do not exceed 1 turn length, and the relative maximum errors are 1.26%, 3.31% and 2.28%. It can be seen that the positioning has a high accuracy under various short-circuit fault types.
[0111] Another aspect of the present invention provides an offline diagnostic device for rotor winding short-circuit fault of a variable speed pumped storage unit, comprising:
[0112] The pulse generation module is used to generate three low-voltage pulse signals at the same time. The pulse signals are generated using the Heidler function model, and the generation of waveforms with different steepness can be achieved by adjusting the wave head time constant τ1 and the wave tail time constant τ2;
[0113] Impedance matching module, used to suppress the reflection of traveling waves at the end of the rotor winding to interfere with the fault signal characteristics, and the impedance matching effect can be achieved by adjusting the sliding rheostat;
[0114] Signal acquisition module, used to collect and display three-way voltage signals, and can store corresponding collected data;
[0115] The data analysis and calculation module obtains the data collected by the signal acquisition module and then processes and analyzes the collected data to implement the above steps.
[0116] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for offline diagnosis of rotor winding short-circuit fault of a variable speed pumped storage unit, wherein the variable speed pumped storage unit shuts down after a rotor winding short-circuit fault occurs, characterized in that: The following steps are involved: S1. Simultaneously inject the same low-voltage pulse signal into the three-phase windings of the rotor and collect the three-phase feedback voltage signal, recorded as u A 、u B 、u C , make a difference between the three-phase voltage signals and get three characteristic waveforms u AB 、u BC 、u CA ; S2. Determine whether the three characteristic waveforms are activated respectively, and determine whether the rotor winding short-circuit fault type is a phase-to-phase short circuit or a turn-to-turn short circuit based on the number of activated characteristic waveforms; If it is a turn-to-turn short circuit fault, go to step S3; if it is a phase-to-phase short circuit fault, go to step S4; S3. First, use the characteristics of the activation characteristic waveform to determine the fault phase where the turn-to-turn short circuit occurs, and then select the fault phase as the characteristic waveform formed by the subtracted phase, and further determine whether the turn-to-turn short circuit fault type is the same branch turn-to-turn short circuit or the same phase different branch turn-to-turn short circuit, and proceed to step S5; S4. Using the characteristics of the activated characteristic waveform to determine the faulty two phases when a phase-to-phase short circuit occurs; S5. Select the corresponding fault phase voltage acquisition signal and characteristic waveform according to the fault type and fault phase, locate the time corresponding to the starting mutation position of the voltage signal and characteristic waveform, and determine the fault position through the time difference between the two and the measured wave speed.
2. The method for off-line diagnosis of rotor winding short-circuit fault of variable speed pumped storage unit according to claim 1, characterized in that: The step S2 specifically includes: S201. Determine whether the maximum absolute value of the three characteristic waveforms is greater than the set value. If it is greater than the set value, it is defined as the characteristic waveform is activated, otherwise it is defined as the characteristic waveform is not activated; S202. If all three characteristic waveforms are activated, the fault type is determined to be a phase-to-phase short circuit; if only two characteristic waveforms are activated, the fault type is determined to be a turn-to-turn short circuit.
3. The method for off-line diagnosis of rotor winding short-circuit fault of variable speed pumped storage unit according to claim 1, characterized in that: The step S3 specifically includes: S301. The phases commonly contained in the two characteristic waveforms activated at the same time are determined to be fault phases; S302. Select the fault phase as the characteristic waveform formed by the subtracted phase; S303. Use the maximum peak time difference detection method to determine the type of inter-turn short circuit fault. The maximum peak time difference detection method is: Among them, T is the propagation time of the traveling wave on the length of the two branches of the rotor winding, which serves as the threshold; t0 is the injection pulse positioning time; u is the selected characteristic waveform; t(max(u)) is the time corresponding to the positive maximum value on the selected characteristic waveform.
4. The method for off-line diagnosis of rotor winding short-circuit fault of variable speed pumped storage unit according to claim 1, characterized in that: The step S4 specifically includes: When the fault type is a phase-to-phase short circuit fault, the two phases corresponding to the largest amplitudes among the three characteristic waveforms are identified as the two faulty phases.
5. The method for off-line diagnosis of rotor winding short-circuit fault of variable speed pumped storage unit according to claim 1, characterized in that: In step S5, if the fault type is a short circuit between turns of the same branch or a short circuit between turns of different branches of the same phase, then one of the fault phase voltage signal and the characteristic waveform containing the fault phase is selected for the next step of fault location identification; if the fault type is a phase-to-phase short circuit, then one of the two fault phase voltage signals and the characteristic waveform composed of the two fault phases are randomly selected for the next step of fault location identification.
6. The variable speed pumped storage unit rotor winding short circuit fault offline diagnosis method according to claim 5, characterized in that: The step S5 specifically includes: S501. Using VMD-TEO method to locate the selected fault phase voltage signal u x The time corresponding to the starting mutation position is the injection pulse time; S502. Using VMD-TEO method to locate the selected characteristic waveform u xy The time corresponding to the starting mutation position is the arrival time of the reflected traveling wave at the fault point, x, y = A, B, C; S503. Determine the distance between the fault point and the rotor winding head end by using the difference between the injection pulse time and the arrival time of the reflected traveling wave at the fault point and the measured wave speed, which is Where v is the measured wave velocity, and Δt is the time difference between the injection pulse and the arrival time of the reflected traveling wave at the fault point.
7. The method for off-line diagnosis of rotor winding short-circuit fault of variable speed pumped storage unit according to claim 6, characterized in that: The measured wave velocity is obtained by the following method: The low-voltage pulse signal is injected into only one phase through the collector ring of the rotor winding, and the pulse signal is received in any of the other two phases. The difference between the injection signal and the reception signal time is the propagation time of the traveling wave on the length of the two branches of the rotor winding. The measured wave speed can be obtained as: Wherein, L is the length of a single branch of the rotor winding, t0 is the injection pulse positioning time, and t1 is the receiving pulse positioning time of the other two phases.
8. An off-line diagnostic device for rotor winding short-circuit fault of a variable speed pumped storage unit, characterized in that: include: A pulse generation module is used to generate three low-voltage pulse signals simultaneously; Impedance matching module, used to suppress the reflection of traveling waves at the end of the rotor winding and interfere with the fault signal characteristics; Signal acquisition module, used to collect and display three-way voltage signals, and can store corresponding collected data; The data analysis and calculation module is used to obtain the data collected by the signal acquisition module and then process and analyze the collected data to implement the method in any one of claims 1 to 7.
9. The variable speed pumped storage unit rotor winding short circuit fault offline diagnosis device as claimed in claim 8, characterized in that: The pulse signal is generated using a Heidler function model.
10. The variable speed pumped storage unit rotor winding short circuit fault offline diagnosis device according to claim 8, characterized in that: The impedance matching module achieves impedance matching effect by adjusting the sliding rheostat.
