Offshore doubly-fed motor rotor winding turn-to-turn short circuit fault identification method based on magnetic field vibration

Abstract

The invention relates to an offshore doubly-fed motor rotor winding turn-to-turn short circuit fault identification method based on magnetic field oscillation. The method comprises the following steps: S1, obtaining a three-phase stator end line voltage, a three-phase rotor end phase current and a rotating speed of an offshore doubly-fed motor to be subjected to fault identification; s2, obtaining a radius and a swing angle required by the characteristic quantity polar coordinate image; s3, obtaining a quadrature-direct axis current residual error and ideal quadrature-direct axis current; s4, calculating to obtain a fault severity factor, judging whether the fault severity factor reaches a pre-configured severity factor threshold value or not, if so, executing the step S5, and if not, executing the step S5; s5, obtaining a quadrature-direct axis current signal difference value, and calculating a fault current initial phase angle based on the quadrature-direct axis current signal difference value; and S6, judging a fault phase according to the absolute value of the difference between the initial phase angle of the fault current and the initial phase angle of the phase current of the three-phase rotor end. Compared with the prior art, the method has the advantages of reducing the missed judgment rate and the misjudgment rate and the like.

Description

Technical Field

[0001] The present invention relates to the field of motor fault identification, and in particular to a method for identifying inter-turn short-circuit faults of rotor windings of an offshore doubly-fed motor based on magnetic field oscillation. Background Art

[0002] Offshore wind power has experienced rapid growth in recent years, driven by advantages such as high efficiency, stable wind speeds, and minimal land occupation. Cumulative installed capacity continues to rise. Doubly-fed wind turbine systems, whose inverters only carry slip power, offer advantages such as low capacity and minimal investment losses. These systems became a mainstream model in the early stages of offshore wind power development. However, with the rapid development of offshore wind power, over 60,000 units, many of which are doubly-fed, have entered their peak warranty period. Failures can significantly impact the safe and stable operation of the power grid. Prompt detection of faults is crucial to prevent rapid deterioration, resulting in further economic losses or casualties.

[0003] Among the major components of offshore doubly-fed (DFIG) generators, faults in equipment like generators and gearboxes result in long downtimes, far exceeding those of other components. Among all generator failures, rotor winding failures have garnered significant attention due to the significant engineering losses they cause. While research on rotor interturn short circuits has increased in recent years, the following challenges remain: 1) The fault signal of a rotor interturn short circuit is much weaker than that of a stator interturn short circuit due to the slip of the DFIG; 2) motor parameters can easily shift after a fault and mismatch the values on the nameplate; 3) the nonlinear control of the rotor-side inverter results in high harmonic content in the rotor voltage, making traditional stator interturn short circuit methods inadequate. These factors complicate the identification of interturn short circuits in DFIG rotor windings. During operation, rotor windings are susceptible to interturn insulation loss due to manufacturing process limitations and wear caused by high-speed rotation, which can lead to rotor interturn short circuits. It may also evolve into a phase-to-phase short circuit in a very short time, or even cause irreversible faults such as ground short circuit, causing damage to the rotor core and posing a huge threat to the safe operation of the unit. There is an urgent need to propose new fault characteristics that meet engineering needs.

[0004] Therefore, in order to detect electrical faults in offshore wind turbines as early as possible and avoid devastating accidents caused by missed faults in the early stages, scholars such as Li Yonggang and Du Wei have proposed using several motor rotor inter-turn short-circuit characteristic quantities, including vibration, harmonics, and stray magnetic flux, to diagnose the occurrence of faults. However, these characteristics are applicable to synchronous or asynchronous motors. Due to the influence of nonlinear control of the rotor-side inverter, these methods are not fully applicable to doubly fed motors. In addition, some characteristics require additional hardware equipment, which increases the diagnostic cost. Li Junqing et al. published a paper on stator-side current harmonic analysis of doubly fed induction generators during rotor inter-turn short-circuit in the Automation of Power Systems. By detecting the harmonic characteristics on the stator side, early diagnosis of rotor inter-turn short-circuit is achieved. However, due to the influence of slip rate, the originally weak rotor fault characteristics are further reduced, making it easy to miss or misjudge. Wei Shurong, Ren Zixu, et al. published a paper titled "Early Fault Identification of Inter-Turn Short Circuit in Offshore Doubly-Fed Wind Turbine Rotor Windings Based on Bilateral Flux Observation Difference" in the Proceedings of the Chinese Society of Electrical Engineering. The paper proposed a method for diagnosing rotor inter-turn short circuit by using bilateral flux observation difference, which solved the problem of weak rotor fault characteristics caused by slip rate. However, when there is an error in the speed measurement, the size of the characteristic quantity overlaps in different degrees of inter-turn short circuit, which can easily lead to misjudgment. Summary of the Invention

[0005] The purpose of the present invention is to solve the above-mentioned defects of the prior art and provide a method for identifying inter-turn short-circuit faults of offshore doubly-fed generator rotor windings based on magnetic field oscillation.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A method for identifying inter-turn short-circuit faults in rotor windings of an offshore doubly-fed generator based on magnetic field oscillation comprises:

[0008] Step S1: obtaining the three-phase stator terminal line voltage, three-phase rotor terminal phase current and speed of the offshore doubly-fed generator to be fault-identified;

[0009] Step S2: Substitute the three-phase stator terminal line voltage, the three-phase rotor terminal phase current and the speed into the magnetic field swing characteristic quantity formula to obtain the radius and swing angle required for the characteristic quantity polar coordinate image;

[0010] Step S3: Based on the three-phase stator terminal line voltage and the three-phase rotor terminal phase current, the AC-axis voltage and the AC-axis current are obtained and used as the input of the Romberg observer to obtain the AC-axis current residual and the ideal AC-axis current;

[0011] Step S4: Calculate the fault severity factor based on the ideal quadrature and direct-axis currents, the swing angle range, and the pre-configured fault quadrature and direct-axis currents, and determine whether the fault severity factor reaches a pre-configured severity factor threshold. If so, proceed to step S5.

[0012] Step S5: subtracting the motor rotor quadrature and direct-axis currents filtered by sliding average before the fault of the motor to be identified from the pre-configured fault quadrature and direct-axis currents to obtain a quadrature and direct-axis current signal difference, and calculating the fault current initial phase angle based on the quadrature and direct-axis current signal difference;

[0013] Step S6: Determine the fault phase according to the absolute value of the difference between the initial phase angle of the fault current and the initial phase angle of the three-phase rotor terminal phase current.

[0014] The radius and swing angle required for the characteristic polar coordinate image are:

[0015]

[0016] Where: r(t) is the radius required for the polar coordinate image of the feature quantity, is the rotor current space vector, abs(·) is the absolute value, Re(·) is the real part of the vector, δ(t) is the swing angle required for the polar coordinate image of the characteristic quantity, w r is the rotation speed, is the stator voltage space vector.

[0017] The rotor current space vector is:

[0018]

[0019] The stator voltage space vector is:

[0020]

[0021] Where: AB (t), ν BC (t), ν CA (t) is the three-phase stator terminal line voltage, i a (t), i b (t), i c (t) is the three-phase rotor terminal phase current.

[0022] The step S3 comprises:

[0023] Step S3-1: transform the three-phase stator terminal line voltage and the three-phase rotor terminal phase current into the direct-quadrature axis coordinate system rotating in the positive direction at the slip angular velocity to obtain the direct-quadrature axis voltage u qr 、u dr and the DC axis current i qr 、i dr 、i qs 、i ds ;

[0024] Step S3-2: Using the obtained AC-axis voltage and AC-axis current as inputs of the Romberg observer to obtain the AC-axis current residual and the ideal AC-axis current.

[0025] The mathematical expression of the Lumberg observer is:

[0026]

[0027] in: is the estimated value of the quadrature and direct axis current residual, is the estimated value of the ideal AC and DC axis current, L r To convert the rotor self-inductance to the direct-direct axis, L m To convert the stator-rotor mutual inductance to the direct-quasi-direct axis, R r is the rotor resistance, ω s is the slip angular velocity, k1 to k8 are feedback coefficients.

[0028] The fault severity factor is:

[0029]

[0030] Δδ=δ max (t)-δ min (t)

[0031] Where: FI is the fault severity factor, Δδ is the swing angle range, i' dr_dc 、i' qr_dc is the pre-configured fault AC and DC axis current, δ max (t) is the maximum swing angle, δ min (t) is the minimum swing angle.

[0032] The pre-configured fault AC and DC axis currents are obtained based on the following method:

[0033] The AC and DC axis currents of the motor are filtered through a sliding average, and the mean value after the rotor winding inter-turn short circuit fault is obtained as the pre-configured fault AC and DC axis current.

[0034] The initial phase angle of the fault current is:

[0035]

[0036] Where: θ f is the initial phase angle of the fault current, Δi dr_dc , Δi qr_dc is the difference between the quadrature and direct axis current signals.

[0037] The step S6 comprises:

[0038] Step S6-1: Calculate the absolute value of the difference between the initial phase angle of the fault current and the initial phase angle of the three-phase rotor terminal phase current as the fault location parameter:

[0039] Step S6-2: Determine which phase fault location parameter is closest to 0, and determine that a rotor winding inter-turn short circuit fault exists in the corresponding phase.

[0040] A device for identifying inter-turn short-circuit faults of rotor windings of an offshore doubly-fed generator based on magnetic field oscillation comprises a memory, a processor, and a program stored in the memory, wherein the processor implements the above-mentioned method when executing the program.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. The acquired three-phase stator terminal line voltage, three-phase rotor terminal phase current, and motor speed are reconstructed into radius and swing angle, and the magnetic field swing image drawn in polar coordinates based on the radius and swing angle is used as the fault feature. This feature can accurately identify the inter-turn short-circuit fault of the doubly fed motor rotor winding and is not affected by the nonlinear control of the rotor-side inverter. Combined with the residual model of the inter-turn short-circuit current of the doubly fed motor rotor, a quantitative severity index and a fault phase location index that are not affected by slip rate changes and parameter mismatch are proposed, which solves the fuzzy judgment of image recognition. The entire process does not require additional hardware equipment and can realize online diagnosis so that timely processing can be made, thereby extending the service life of the doubly fed motor, avoiding catastrophic failures, and reducing the economic losses caused by failures.

[0043] 2. Through specially designed FI and fault location parameters, the problem of missed judgment and misjudgment caused by the further reduction of the originally weak rotor fault characteristics due to the influence of slip rate is solved. The problem of misjudgment caused by the overlap of the characteristic values in different degrees of inter-turn short circuit when there is an error in the speed measurement is also solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a simulation result diagram of the magnetic field swing and quantitative positioning index of the doubly-fed generator before and after a fault of the present invention;

[0045] Figure 2 Schematic diagram of the main steps of the method of the present invention;

[0046] Figure 3 This is a diagram showing the results of a magnetic field oscillation experiment before and after a fault of the doubly-fed generator of the present invention;

[0047] Figure 4 Graph showing the magnetic field oscillation test results of the doubly-fed generator of the present invention at 1.43% and 3.57% fault levels;

[0048] Figure 5 This is a simulation result diagram of magnetic field oscillation and quantitative positioning index under parameter mismatch of the doubly-fed generator of the present invention;

[0049] Figure 6This is a simulation result diagram of the magnetic field oscillation and quantitative positioning index of the doubly-fed generator under speed changes of the present invention. DETAILED DESCRIPTION

[0050] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0051] The three-phase stator voltage, rotor current, and motor speed of the doubly-fed generator are obtained and reconstructed into the radius and swing angle in polar coordinates. The polar coordinate image drawn with the radius r(t) and the swing angle δ(t) is used as the magnetic field swing feature. This feature is:

[0052]

[0053] It is a straight line with a continuously changing length. After the fault:

[0054]

[0055] It presents a petal shape, and the fault characteristics are obvious. The Hilbert transform is used to calculate the envelope of the swing angle curve, and then the maximum swing angle Δδ=δ is calculated in real time. max -δ min And stored, due to the ambiguity of image judgment, further quantitative analysis is required.

[0056] There are many methods to establish the fault model of the doubly fed generator, such as the multi-loop method and the finite element method. However, due to its complex structure, it is not conducive to online diagnosis, so it is rarely used in engineering for online diagnosis. The existing mathematical model of the doubly fed asynchronous motor rotor a phase inter-turn short circuit fault in the three-phase stationary coordinate system (abc coordinate system). Transform it to the angular velocity ω s In the forward rotating dq axis coordinate system, the focus is on the rotor winding part, which is improved to the current residual model when the rotor turns are short-circuited.

[0057] like Figure 1 The simulation results of magnetic field swing before and after the double-fed generator fault are shown in the figure. The polar axis is current / A and the vertical axis is angle difference / °. Figure 1 Under the operating condition of the doubly fed generator rotor winding inter-turn short-circuit current residual model, the residual model is:

[0058] u dr_f =R r (i dr +Δi dr )+L r p(i dr +Δi dr )

[0059] -w s Lr (i qr +Δi qr )+L m pi ds -w s L m i qs

[0060] u qr_f =R r (i qr +Δi dr )+L r p(i qr +Δi qr )

[0061] +w s L r (i dr +Δi dr )+L m pi qs +w s L m i ds

[0062] In the formula, p is the differential symbol; ω s is the slip angular velocity; u dr_f 、u qr_f is the DC-axis rotor voltage after the fault; i dr 、i qr is the direct-axis rotor current under normal conditions; Δi dr , Δi qr The residual of the quadrature and direct axis current after the fault; i ds 、i qs is the DC axis stator current; R r is the rotor resistance; L r is the rotor self-inductance converted to the orthogonal axis coordinate system; L m is the mutual inductance of the stator and rotor converted to the orthogonal axis coordinate system.

[0063] Assumption i f =I f sin(θ s +θ f ), after substituting, we can get:

[0064]

[0065] It can be seen that the current residual is transformed into ω s The angular velocity of the coordinate system can be decomposed into a DC quantity and a 2ω sComponents, of which the double frequency component is the cause of magnetic field oscillation. Because the rotor current residual cannot be obtained directly, it is necessary to input the rotor current into the Lumberg observer to separate the ideal dq axis current i dr +ji qr and dq axis current residuals.

[0066] In order to propose appropriate quantitative and positioning indicators, it is necessary to analyze the robustness of the mismatch between the characteristic rotor speed and motor parameters. Assume that the rotor resistance change is R r +ΔR r , the rotor self-inductance changes to L r +ΔL r , the rotor mutual inductance changes to L m +ΔL m , when the motor is healthy, the voltage equation changes to:

[0067]

[0068] When the motor is in a healthy state and the load is fixed, i dr and i qr It can be approximately regarded as a constant. Therefore, the di on the left side of the equation dr / dt and di qr The / dt value is 0 and the left side is a DC quantity. Solving the above differential equation with constant coefficients, the current residual solution that satisfies the above equation is a DC quantity. Therefore, when the motor parameters do not match, only a DC bias will appear in the current residual. s The component is irrelevant. In order to avoid the influence of parameter mismatch, the quantitative index can select information related to the double frequency component. At the same time, considering the robustness of the speed change, the present invention uses a new quantitative index to define the fault severity factor.

[0069]

[0070] The pre-configured fault AC and DC axis currents are obtained based on the following:

[0071] The AC and DC axis currents of the motor are filtered through a sliding average, and the mean value after the rotor winding inter-turn short circuit fault is obtained as the pre-configured fault AC and DC axis current.

[0072] Once the quantitative indicator exceeds the threshold, a fault can be determined, and the fault phase can be further determined. Due to the fact that the initial phase angle of the fault circulating current and the initial phase angle of the fault phase current are approximately equal, it is necessary to calculate the residual DC value of the current related to the fault. This DC value can be calculated as:

[0073]

[0074] Therefore, a method for identifying inter-turn short-circuit faults of offshore doubly-fed generator rotor windings based on magnetic field oscillation is provided. Figure 2 Shown, including:

[0075] Step S1: obtaining the three-phase stator terminal line voltage, three-phase rotor terminal phase current and speed of the offshore doubly-fed generator rotor to be fault-identified;

[0076] Step S2: Substitute the three-phase stator terminal line voltage, the three-phase rotor terminal phase current, and the speed into the magnetic field swing characteristic quantity formula to obtain the radius and swing angle required for the characteristic quantity polar coordinate image:

[0077]

[0078] Where: r(t) is the radius required for the polar coordinate image of the feature quantity, is the rotor current space vector, abs(·) is the absolute value, Re(·) is the real part of the vector, δ(t) is the swing angle required for the polar coordinate image of the characteristic quantity, w r is the rotation speed, is the stator voltage space vector.

[0079] The rotor current space vector is:

[0080]

[0081] The stator voltage space vector is:

[0082]

[0083] Where: AB (t), ν BC (t), ν CA (t) is the three-phase stator terminal line voltage, i a (t), i b (t), i c (t) is the three-phase rotor terminal phase current.

[0084] Step S3: Based on the three-phase stator terminal line voltage and the three-phase rotor terminal phase current, the AC-axis voltage and the AC-axis current are obtained and used as the input of the Romberg observer to obtain the AC-axis current residual and the ideal AC-axis current;

[0085] Step S3 includes:

[0086] Step S3-1: transform the three-phase stator terminal line voltage and the three-phase rotor terminal phase current into the direct-quadrature axis coordinate system rotating in the positive direction at the slip angular velocity to obtain the direct-quadrature axis voltage u qr 、u dr and the DC axis current i qr 、i dr 、i qs 、ids ;

[0087] Step S3-2: Using the obtained AC-axis voltage and AC-axis current as inputs of the Romberg observer to obtain the AC-axis current residual and the ideal AC-axis current.

[0088] The mathematical expression of the Romberg observer is:

[0089]

[0090] in: is the estimated value of the quadrature and direct axis current residual, is the estimated value of the ideal AC and DC axis current, L r To convert the rotor self-inductance to the direct-alternating axis, L m To convert the stator-rotor mutual inductance to the direct-quasi-direct axis, R r is the rotor resistance, ω s is the slip angular velocity, k1 to k8 are feedback coefficients.

[0091] Step S4: Calculate the fault severity factor based on the ideal quadrature and direct-axis currents, the swing angle range, and the pre-configured fault quadrature and direct-axis currents, and determine whether the fault severity factor reaches a pre-configured severity factor threshold. If so, proceed to step S5.

[0092] The fault severity factor is:

[0093]

[0094] Δδ=δ max (t)-δ min (t)

[0095] Where: FI is the fault severity factor, Δδ is the swing angle range, i' dr_dc 、i' qr_dc is the pre-configured fault AC and DC axis current, δ max (t) is the maximum swing angle, δ min (t) is the minimum value of the swing angle.

[0096] Step S5: Subtract the motor rotor quadrature and direct-axis currents filtered by sliding average before the fault of the motor to be identified from the pre-configured fault quadrature and direct-axis currents to obtain the quadrature and direct-axis current signal difference, and calculate the fault current initial phase angle based on the quadrature and direct-axis current signal difference:

[0097]

[0098] Where: θ f is the initial phase angle of the fault current, Δi dr_dc , Δi qr_dc is the difference between the quadrature and direct axis current signals.

[0099] Step S6: judging the fault phase according to the absolute value of the difference between the initial phase angle of the fault current and the initial phase angle of the three-phase rotor terminal phase current, including:

[0100] Step S6-1: Calculate the absolute value of the difference between the initial phase angle of the fault current and the initial phase angle of the three-phase rotor terminal phase current as the fault location parameter:

[0101] Step S6-2: Determine which phase fault location parameter is closest to 0, and determine that a rotor winding inter-turn short circuit fault exists in the corresponding phase.

[0102] For the simulation and experiment of magnetic field oscillation, the working conditions are that the grid voltage is a symmetrical three-phase voltage, the rotor slip is 20%, the sampling frequency fs is 20000hz during signal measurement, the stator active power is set to -3000W (with the power flowing into the motor as the positive direction) during normal operation, and the reactive power is set to 0var. During simulation, the motor short-circuit coefficient changes from 0 to 0.0357 at t=10s, while the same type of fault occurs at around t=6.25s during the experiment. In order to protect the experimental equipment, the system is derated when a fault occurs, that is, when a turn-to-turn short-circuit fault occurs, the stator active power is set to -2700W and the reactive power is set to 0var. The simulation and experimental results are shown in Figure 2. Figure 1 、 Figure 3 shown.

[0103] from Figure 3 As can be seen from the figure, the simulation and experimental results are basically consistent. The magnetic field oscillation image before the fault is a straight line, and after the fault it becomes a petal shape, and the fault characteristics change significantly. Figure 3 In the experiment, the magnetic field oscillation image is not completely a straight line under normal conditions, and the petal shape of the experiment after the fault is slightly larger than that of the simulation. This is due to reasons such as sensor measurement errors under experimental conditions. However, the change patterns of the two are consistent. The simulation and experimental results prove the correctness of the theoretical derivation in the previous article.

[0104] Figure 4 The following are the experimental magnetic field oscillation results for different fault severities. Due to experimental limitations (the rotor only has two short-circuit taps), two fault severities of 1.43% and 3.57% were simulated. It can be seen that the magnetic field oscillation petals gradually increase with the severity of the fault.

[0105] Figure 5 、 Figure 6 This is to illustrate the robustness of the method of the present invention to changes in motor speed and parameter mismatch.

[0106] Figure 5 The active power is set to 3kw and the slip rate is set to 20%. Assume that the parameters change by 50% with the external environment, that is, R r ±0.5R r , L r±0.5L r , L m ±0.5L m It can be seen that the magnetic field oscillation diagram is always a straight line when the motor is healthy. The motor fails at 10s and the short circuit coefficient is 3.57%. It can be seen that the positions of different petals are slightly different after the failure, but the swing angle and radius length are almost unchanged. The quantitative index diagram can be FI and The quantitative index is robust to parameter mismatch. Among the positioning indicators, ga is always the smallest and varies around 0°, indicating that phase a is at fault.

[0107] Figure 6 The active power is set at 3 kW, and the initial slip rate is 20%. Starting from 15 seconds, it increases by 20% every 3 seconds until it reaches 80%. The magnetic field oscillation diagram when the motor is healthy shows that even if the speed changes, the image remains a straight line. The motor fails at 10 seconds, and the short-circuit coefficient is 3.57%. It can be seen that the petal length is increasing, and the swing angle amplitude is almost unchanged. FI remains at around 0.05 at different speeds, which is consistent with theoretical analysis. The positioning indicator shows that even if g a Not exactly 0, but similar to g b 、g c Since the value is always closest to 0, it can be determined that the fault is in phase A.

[0108] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

Claims

1. A method for identifying inter-turn short-circuit faults in offshore doubly-fed generator rotor windings based on magnetic field oscillation, characterized in that: include: Step S1: obtaining the three-phase stator terminal line voltage, three-phase rotor terminal phase current and speed of the offshore doubly-fed generator to be fault-identified; Step S2: Substitute the three-phase stator terminal line voltage, the three-phase rotor terminal phase current and the speed into the magnetic field swing characteristic quantity formula to obtain the radius and swing angle required for the characteristic quantity polar coordinate image; Step S3: Based on the three-phase stator terminal line voltage and the three-phase rotor terminal phase current, the AC-axis voltage and the AC-axis current are obtained and used as the input of the Romberg observer to obtain the AC-axis current residual and the ideal AC-axis current; Step S4: Calculate the fault severity factor based on the ideal quadrature and direct-axis currents, the swing angle range, and the pre-configured fault quadrature and direct-axis currents, and determine whether the fault severity factor reaches a pre-configured severity factor threshold. If so, proceed to step S5. Step S5: subtracting the motor rotor quadrature and direct-axis currents filtered by sliding average before the fault of the motor to be identified from the pre-configured fault quadrature and direct-axis currents to obtain a quadrature and direct-axis current signal difference, and calculating the fault current initial phase angle based on the quadrature and direct-axis current signal difference; Step S6: Determine the fault phase according to the absolute value of the difference between the initial phase angle of the fault current and the initial phase angle of the three-phase rotor terminal phase current.

2. The method for identifying inter-turn short circuit faults of offshore doubly-fed generator rotor windings based on magnetic field oscillation according to claim 1, characterized in that: The radius and swing angle required for the characteristic polar coordinate image are: Where: r(t) is the radius required for the polar coordinate image of the feature quantity, is the rotor current space vector, abs(·) is the absolute value, Re(·) is the real part of the vector, δ(t) is the swing angle required for the polar coordinate image of the characteristic quantity, w r is the rotation speed, is the stator voltage space vector.

3. The method for identifying inter-turn short circuit faults of offshore doubly-fed generator rotor windings based on magnetic field oscillation according to claim 2, characterized in that: The rotor current space vector is: The stator voltage space vector is: Where: AB (t), ν BC (t), ν CA (t) is the three-phase stator terminal line voltage, i a (t), i b (t), i c (t) is the three-phase rotor terminal phase current.

4. A method for identifying inter-turn short circuit faults in rotor windings of an offshore doubly-fed generator based on magnetic field oscillation according to claim 1, characterized in that: The step S3 comprises: Step S3-1: transform the three-phase stator terminal line voltage and the three-phase rotor terminal phase current into the direct-quadrature axis coordinate system rotating in the positive direction at the slip angular velocity to obtain the direct-quadrature axis voltage u qr 、u dr and the direct-axis current i qr 、i dr 、i qs 、i ds ; Step S3-2: Using the obtained AC-axis voltage and AC-axis current as inputs of the Romberg observer to obtain the AC-axis current residual and the ideal AC-axis current.

5. A method for identifying inter-turn short circuit faults in offshore doubly-fed generator rotor windings based on magnetic field oscillation according to claim 4, characterized in that: The mathematical expression of the Lumberg observer is: in: is the estimated value of the quadrature and direct axis current residual, is the estimated value of the ideal AC and DC axis current, L r To convert the rotor self-inductance to the direct-alternating axis, L m To convert the stator-rotor mutual inductance to the direct-quasi-direct axis, R r is the rotor resistance, ω s is the slip angular velocity, k1 to k8 are feedback coefficients.

6. A method for identifying inter-turn short circuit faults in offshore doubly-fed generator rotor windings based on magnetic field oscillation according to claim 5, characterized in that: The fault severity factor is: Δδ=δ max (t)-d min (t) Where: FI is the fault severity factor, Δδ is the swing angle range, i' dr_dc 、i' qr_dc is the pre-configured fault AC and DC axis current, δ max (t) is the maximum swing angle, δ min (t) is the minimum swing angle.

7. A method for identifying inter-turn short circuit faults in offshore doubly-fed generator rotor windings based on magnetic field oscillation according to claim 6, characterized in that: The pre-configured fault AC and DC axis currents are obtained based on the following method: The AC and DC axis currents of the motor are filtered through a sliding average, and the mean value after the rotor winding inter-turn short circuit fault is obtained as the pre-configured fault AC and DC axis current.

8. A method for identifying inter-turn short circuit faults in offshore doubly-fed generator rotor windings based on magnetic field oscillation according to claim 1, characterized in that: The initial phase angle of the fault current is: Where: θ f is the initial phase angle of the fault current, Δi dr_dc , Δi qr_dc is the difference between the quadrature and direct axis current signals.

9. A method for identifying inter-turn short circuit faults in offshore doubly-fed generator rotor windings based on magnetic field oscillation according to claim 8, characterized in that: The step S6 comprises: Step S6-1: Calculate the absolute value of the difference between the initial phase angle of the fault current and the initial phase angle of the three-phase rotor terminal phase current as the fault location parameter: Step S6-2: Determine which phase fault location parameter is closest to 0, and determine that a rotor winding inter-turn short circuit fault exists in the corresponding phase.

10. A device for identifying inter-turn short-circuit faults in rotor windings of an offshore doubly-fed generator based on magnetic field oscillation, comprising a memory, a processor, and a program stored in the memory, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 9 is implemented.

Images