CN108196187B - Fault Diagnosis Method for Three-phase Asymmetric Loads of Rotor Windings of Doubly-fed Wind Turbines - Google Patents

Abstract

The invention discloses a three-phase asymmetric load fault diagnosis method for a rotor winding of a doubly-fed wind turbine. Step 1: build a rotor winding fault simulation system of a doubly-fed wind turbine on a dynamic model experimental platform; step 2: in the motor running In different states, the rotor current or voltage signal before and after the fault on the rotor side is obtained, and the fault recording and analysis device performs corresponding recording, reads the data through MATLAB, and performs DC component filtering and FFT transformation processing; Step 3: According to the change of amplitude The frequency that changes significantly before and after the fault is extracted as the characteristic frequency for diagnosing the fault, which is used to judge whether the three-phase asymmetric load fault occurs in the rotor winding of the doubly-fed wind turbine. The diagnosis result of the invention is very close to the real value, the diagnosis result is effective, the diagnosis accuracy is high, and the operability is strong.

Description

Fault Diagnosis Method for Three-phase Asymmetric Loads of Rotor Windings of Doubly-fed Wind Turbines

technical field

The invention relates to a fault diagnosis method for a three-phase asymmetric load of a rotor winding of a double-fed wind turbine generator, and belongs to the technical field of drive motor state detection and fault diagnosis.

Background technique

The normal operation of the generator is of great significance to the safety and stability of the entire power grid. In wind farms, double-fed wind turbines have become the mainstream models due to their high efficiency, low cost and adjustable power factor. Due to the harsh operating environment, wind turbines cause far more failures than conventional generators. Therefore, in order to enhance the reliability of wind turbines, as well as to carry out early warning of failures and timely maintenance, the research on the online monitoring technology of wind turbines is very important.

Due to its external circuit and the connection of a back-to-back converter, the rotor winding of the doubly-fed wind turbine is difficult to detect its operating state, and the parameter adjustment of the generator rotor side controller lags behind, and the active power, reactive power, stator and rotor current and speed are The oscillation before and after the fault is serious, which has a great impact on the motor itself and the power grid. Therefore, it is necessary to strengthen the online monitoring of the rotor winding status information.

At present, domestic and foreign research on the diagnosis of rotor winding faults of DFIG is still in its infancy. The domestic research is relatively mature, and the literature can select the appropriate characteristic signal to find out the fault through the corresponding characteristic method; Yazidi A, a scholar from the University of Picardie in France, builds a model of the normal circuit of the doubly-fed wind turbine and the short circuit between the turns of the rotor winding for simulation research; Scholar Dinkhauser V of the University of Kiel in Germany established a doubly-fed machine model using MATLAB software, introduced the factor u to characterize the severity of inter-turn faults, and used wavelet analysis and Lomborg observer to find fault characteristics; Gritli Y, a scholar from the University of Bologna in Italy, analyzed the rotor windings. The experimental data at the time of failure is analyzed, the wavelet transform is used to quantify the failure, and a fixed threshold is set to identify the failure, but the influence of different control strategies of the rotor converter on the extraction of characteristic signals is not considered.

SUMMARY OF THE INVENTION

Objective: In order to overcome the deficiencies in the prior art, the present invention provides a fault diagnosis method for the three-phase asymmetric load of the rotor winding of a doubly-fed wind turbine.

Technical scheme: in order to solve the above-mentioned technical problems, the technical scheme adopted in the present invention is:

A method for diagnosing a three-phase asymmetric load fault of a rotor winding of a doubly-fed wind turbine generator, comprising the following steps:

Step 1: Build a DFIG rotor winding fault simulation system on the dynamic model experimental platform;

Step 2: Set the different states of motor operation by changing the wind speed and the slip rate. Based on the simulation system, obtain the rotor current or voltage signal before and after the simulated three-phase load asymmetric operation on the rotor side, and carry out the rotor current or voltage signal under different states. Correction, and then use the power fault recording and analysis device to record the corresponding waves, read the data through MATLAB, and perform DC component filtering and FFT transformation processing;

Step 3: Analyze the rotor current or voltage spectrum under different operating states under normal and fault states, and obtain the variation amplitude of ksf and its side frequency, k=2n+1, n is a natural number starting from 1, s is the slip rate, f is the fundamental frequency, and the frequency with obvious changes before and after the fault is extracted according to the rate of change of the amplitude as the characteristic frequency for diagnosing the fault, which is used to judge whether the three-phase asymmetric load fault occurs in the rotor winding of the doubly-fed wind turbine. .

As a preferred solution, the step 1 includes:

Step 1.1: Directly connect the 5.5kW DC motor with the 5.5kW wound asynchronous motor to form a double-fed wind turbine. The DC motor provides external excitation for the asynchronous motor, and the excitation size is controlled and displayed in real time through the excitation control screen of the double-fed fan; The DC motor is directly connected to the output terminal of the wind turbine characteristic simulation device, and receives the adjustment signal of the variable wind speed in real time; the output terminal of the rotor side of the asynchronous motor is connected to the rotor side DSP controller to realize the synchronous, super-synchronous and sub-synchronous state transition of the doubly-fed asynchronous motor, and the other side The output end is directly connected to the load and the power fault recording and analysis device, and finally connected to the DSP controller on the grid side to realize power transmission with the grid; the load is simulated by a light bulb.

As a preferred solution, the step 1 also includes:

Step 1.2: Test the performance of the voltage and current transformers before the experiment, verify the voltage transformers through the power frequency power supply, voltage regulator, and multimeter, and the secondary side cannot be short-circuited, string resistors in the loop, and use voltage regulators and clamp-type ammeters. Verify that the current transformer and the secondary side cannot be opened;

Step 1.3: When collecting voltage and current signals, in order to avoid the phenomenon of DC magnetization caused by low-frequency signals, and the phenomenon of waveform distortion on the secondary side caused by exceeding the frequency transmission range of the power frequency transformer, the Hall current sensor with the model HZIA-C06 is selected; Due to the small value of the rotor voltage, the voltage sensor is directly connected to the low-frequency signal measurement device customized by Zhongyuan Huadian;

Step 1.4: In order to avoid the phenomenon of zero point drift due to the absence of zero line on the rotor side, short-circuit any phase with the zero line, and then measure the phase-to-phase voltage or current value of any two phases as the correction value, and perform the measurement on the measured signal. Zero drift correction.

As a preferred solution, the step 2 includes:

Step 2.1: Adjust the motor speed through the wind turbine characteristic simulation device to change the wind speed, thereby changing the slip rate s to set the doubly-fed wind turbine to run in three different operating states, including synchronous state, sub-synchronous state and super-synchronous state. ;

Step 2.2: Select any operating state of the DFIG, when the DFIG works in the normal state, obtain the actual measurement of the rotor current or voltage at the measurement point between the asynchronous motor and the converter through the current or voltage transformer Then send the signal waveform of the measured value minus the correction value to the wave recorder analysis device respectively; filter the signal waveform through the DC component and FFT transformation processing, and obtain the four frequencies of sf, 3sf, 2-3sf, 2+3sf respectively. The normal state amplitude of ;

Step 2.3: Open a certain phase bulb on the simulation system to simulate the three-phase asymmetrical operation fault of the load, obtain the measured value of the rotor current or voltage at the measuring point between the asynchronous motor and the converter again, and then subtract the corrected value from the measured value The signal waveforms are sent to the wave recorder analysis device respectively, and the signal waveforms are filtered by the DC component and processed by FFT transformation, and the fault state amplitudes of the four frequencies of sf, 3sf, 2-3sf, and 2+3sf are obtained respectively.

As a preferred solution, the step 3 includes:

Step 3.1: Calculate the difference between the normal state amplitude and the fault state amplitude of the frequency measurement points sf, 3sf, 2-3sf, 2+3sf between the asynchronous motor and the converter under different operating states of the DFIG. Change value, and then divide the change value by the normal state amplitude to get the change rate;

Step 3.2: Select the frequency value with the largest rate of change among the frequencies of sf, 3sf, 2-3sf, and 2+3sf as the characteristic frequency for diagnosing three-phase asymmetric load faults.

As a preferred solution, the power failure recording and analysis device adopts the customized ZH-2B recording and analysis device of Zhongyuan Huadian.

As a preferred solution, the step 3 also includes:

Take 3sf as the general state characteristic frequency of the doubly-fed wind turbine, 2-3sf as the sub-synchronous state characteristic frequency of the doubly-fed wind turbine, and 2+3sf as the super-synchronous state characteristic frequency of the doubly-fed wind turbine.

Beneficial effects: The method for diagnosing the three-phase asymmetric load of the rotor winding of the doubly-fed wind turbine provided by the present invention has the diagnosis result very close to the real value, the diagnosis result is effective, the diagnosis accuracy is high, and the operability is strong.

In the present invention, under different operating states of the doubly-fed wind turbine, it is determined whether the three-phase load asymmetrical operation fault occurs in the rotor winding of the doubly-fed wind turbine according to the fault characteristic frequency 3sf and the side frequency (2±3sf) in the rotor current signal. and diagnosis. Its advantages are: 1. There is no need to separate the stator and rotor of the doubly-fed wind turbine for separate research, which ensures the integrity of the overall structure of the motor and the reliability of the diagnosis results; 2. Diagnoses the failure of the three-phase load asymmetric operation of the rotor winding in the early stage High sensitivity for early fault diagnosis and containment.

Description of drawings

Fig. 1 is the structure schematic diagram of the rotor winding fault simulation system of the doubly-fed wind generator involved in the present invention;

Fig. 2 is the structure schematic diagram of the double-fed wind turbine generator set involved in the present invention;

Fig. 3 is the fault diagnosis flow chart of three-phase asymmetric load operation of the present invention;

Fig. 4 is the rotor current spectrum diagram of the measurement point on the rotor side of the converter before and after the failure of the three-phase load asymmetric operation of the rotor winding under the sub-synchronous state when s=0.12 involved in the present invention;

5 is a rotor current spectrum diagram of the rotor winding at the rotor side of the converter before and after the failure of the three-phase load asymmetric operation of the rotor winding in the supersynchronous state when s=-0.12 involved in the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

As shown in Figure 1, the parameters in the figure are all rated operating parameters. The wind turbine characteristic simulation device is used to adjust the motor speed, and the excitation control panel of the doubly-fed fan is used to adjust the start and stop of the DSP controller on the rotor side and the grid side, as well as the parameter changes and the real-time display of the speed, active power, reactive power and other values. . Considering the safety and economy of the experiment, replace the load with a light bulb and simulate the load asymmetry by breaking a certain phase of the light bulb;

As shown in Figure 2, the distribution of measuring points is marked with circles in the figure. Among them, the positions of 1-4 points are: the rotor side 1 of the converter, the stator side 2, the rear grid side 3 of the reactor and the rear grid side 4 of the converter. Install appropriate voltage and current transformers in these four places to measure the signal values before and after the fault respectively for subsequent spectrum analysis;

As shown in Figure 3, a method for diagnosing a three-phase asymmetric load fault of a doubly-fed wind turbine rotor winding, the specific operation process is as follows:

Step 1: Build a DFIG rotor winding fault simulation system on the dynamic model experimental platform. The system includes: 5.5kW DC motor, 5.5kW wound asynchronous motor, wind turbine characteristic simulation device, and DFIG excitation control panel , load and power failure recording wave analysis device, rotor side DSP controller, grid side DSP controller.

Considering the safety and economy of the experiment comprehensively, the load is replaced with a light bulb, and the load asymmetry is simulated by breaking a certain phase light bulb; symmetry problem;

Step 2: Set the different states of motor operation by changing the wind speed and the slip rate. Based on the simulation system, obtain the rotor current or voltage signal before and after the simulated three-phase load asymmetric operation on the rotor side, and carry out the rotor current or voltage signal under different states. Correction, and then use the power failure recording and analysis device to record the corresponding waves, the data is stored in the comtrade format, and the data is read through MATLAB, and the DC component filtering and FFT transformation processing are performed;

Step 3: Analyze the rotor current or voltage spectrograms in different operating states under normal and fault states, and obtain ksf (k=2n+1, n is a natural number starting from 1), and the variation amplitude of its side frequency , according to the change rate of the amplitude, the frequency with obvious change before and after the fault is extracted as the characteristic frequency for diagnosing the fault, which is used to judge whether the three-phase asymmetric load fault occurs in the rotor winding of the doubly-fed wind turbine.

The step 1 includes:

Step 1.1: Directly connect the 5.5kW DC motor with the 5.5kW wound asynchronous motor to form a double-fed wind turbine. The DC motor provides external excitation for the asynchronous motor, and the excitation size is controlled and displayed in real time through the excitation control screen of the double-fed fan; The DC motor is directly connected to the output terminal of the wind turbine characteristic simulation device, and receives the adjustment signal of the variable wind speed in real time; the output terminal of the rotor side of the asynchronous motor is connected to the rotor side DSP controller to realize the synchronous, super-synchronous and sub-synchronous state transition of the doubly-fed asynchronous motor, and the other side The output end is directly connected to the load and the power failure recording and analysis device, and finally connected to the DSP controller on the grid side to realize the power transmission with the grid. When building a DFIG rotor winding fault simulation system, the wind turbine characteristic simulation device is used to adjust the motor speed, and the DFIG excitation control panel is used to adjust the start and stop of the rotor side and grid side DSP controllers, as well as parameter changes and speed, active power Real-time display of power and reactive power values; simulate the three-phase asymmetrical operation fault of the load by breaking a certain phase bulb;

Step 1.2: Test the performance of the voltage and current transformers before the experiment, verify the voltage transformers through the power frequency power supply, voltage regulator, multimeter, etc. and the secondary side cannot be short-circuited, string resistance in the loop, use voltage regulators, clamps Verify the current transformer with an ammeter, etc. and the secondary side cannot be opened;

Step 1.3: When collecting voltage and current signals, in order to avoid the phenomenon of DC magnetization caused by low-frequency signals, and the phenomenon of waveform distortion on the secondary side caused by exceeding the frequency transmission range of the power frequency transformer, the Hall current sensor with the model HZIA-C06 is selected. Due to the small value of the rotor voltage, the voltage sensor is directly connected to the low-frequency signal measurement device customized by Zhongyuan Huadian;

Step 1.4: In order to avoid the phenomenon of zero point drift due to the absence of zero line on the rotor side, short-circuit any phase with the zero line, and then measure the phase-to-phase voltage or current value of any two phases as the correction value, and perform the measurement on the measured signal. Zero drift correction.

The step 2 includes:

Step 2.1: Adjust the motor speed through the wind turbine characteristic simulation device to change the wind speed, thereby changing the slip rate s to set the doubly-fed wind turbine to run in three different operating states, including synchronous state, sub-synchronous state and super-synchronous state. .

Step 2.2: Select any operating state of the DFIG. When the DFIG works in the normal state, obtain the asynchronous motor and converter, asynchronous motor and load, transformer and load respectively through the current or voltage transformer. , The measured values of rotor current or voltage at four measurement points between the converter and the transformer, and then send the signal waveforms of the four groups of measured values minus the correction value to the wave recorder analysis device respectively. The four groups of signal waveforms are filtered by DC component and processed by FFT, and the normal state amplitudes of sf, 3sf, 2-3sf, and 2+3sf are obtained respectively; in the experiment, the model customized by Zhongyuan Huadian is ZH-2B. The power fault recording and analysis device records and monitors the operating status of the double-fed wind turbine.

Step 2.3: Open a certain phase bulb on the simulation system to simulate the three-phase asymmetric operation fault of the load, obtain the measured values of the rotor current or voltage at the four measurement points again, and then send the four groups of measured values minus the correction value. To the wave recorder analysis device, the data of the wave recorder analysis device is stored in the comtrade format, the data is read through MATLAB, and the four groups of signal waveforms are filtered by the DC component and processed by FFT transformation to obtain sf, 3sf, 2-3sf, 2+ respectively. 3sf fault state amplitude around the frequency;

The step 3 includes:

Step 3.1: Obtain four sets of change values between the normal state amplitude and the fault state amplitude of the four measurement points sf, 3sf, 2-3sf, 2+3sf under different operating states of the DFIG, and then The change value is divided by the normal state amplitude respectively to obtain the change rate.

Step 3.2: Select the frequency value with the largest rate of change among the frequencies of sf, 3sf, 2-3sf, and 2+3sf in the measurement points on the rotor side of the converter as the characteristic frequency for diagnosing three-phase asymmetric load faults.

In the experimental statistics of different situations, when diagnosing the three-phase load asymmetrical operation fault of the DFIG, 3sf is used as the general state characteristic frequency of the DFIG, and (2-3sf) is used as the DFIG. The characteristic frequency of the sub-synchronous state, (2+3sf) is taken as the characteristic frequency of the super-synchronous state of the doubly-fed wind turbine.

As shown in Figure 4 and Figure 5, the rotor current spectrum diagrams before and after the asymmetrical operation fault of the rotor winding three-phase load in the sub-synchronous state when s=0.12 and the three-phase load of the rotor winding in the super-synchronous state when s=-0.12 are respectively Spectrograms of rotor current before and after the symmetrical running fault. According to the harmonic theory, when the rotor winding of the doubly-fed wind turbine fails, the fault characteristic frequency of the rotor side electrical quantity includes the ksf component (k=3, 5, 7...). Based on this analysis, the different motors obtained in Fig. 4 and Fig. 5 The rotor current spectrogram in the running state, obtain the change amplitude of ksf and its side frequency, and organize the frequency amplitude with obvious change into Table 1.

It can be seen from Table 1 that the fault characteristic frequency amplitude of the rotor-side electrical quantity increases when a fault occurs, and the frequency amplitude at 3sf changes most obviously; The change rate of frequency (2+3sf) is larger than that of (2+3sf), and the fault diagnosis accuracy is higher; when the motor runs in the super-synchronous state, the change rate of frequency (2+3sf) is larger than that of (2-3sf), and the fault diagnosis accuracy is higher. . Therefore, when diagnosing the three-phase asymmetric load operation fault of the rotor winding of the doubly-fed wind turbine, 3sf can be used as the general state characteristic frequency of the motor, (2-3sf) can be used as the subsynchronous state characteristic frequency of the motor, and (2+3sf) can be used as The characteristic frequency of the motor over-synchronization state, so as to further improve the diagnosis accuracy of the fault.

Table 1 is the different frequency changes before and after the rotor winding three-phase asymmetric load operation fault under the sub-synchronization and super-synchronization states involved in the present invention;

The above is only the preferred embodiment of the present invention, it should be pointed out that: for those skilled in the art, without departing from the principle of the present invention, several improvements and modifications can also be made, and these improvements and modifications are also It should be regarded as the protection scope of the present invention.

Claims (1)

1. A fault diagnosis method for three-phase asymmetric loads of a rotor winding of a doubly-fed wind generator is characterized by comprising the following steps of: the method comprises the following steps:

step 1: building a double-fed wind driven generator rotor winding fault simulation system on a moving die experiment platform;

step 2: setting different running states of the motor by changing wind speed and changing slip ratio, acquiring rotor current or voltage signals before and after a simulated three-phase load asymmetric running fault on a rotor side model based on a simulation system, correcting the rotor current or voltage signals in different states, performing corresponding wave recording by using a power fault wave recording analysis device, reading data by using MATLAB, and performing direct current component filtering and FFT conversion processing;

and step 3: analyzing rotor current or voltage spectrograms in different running states under a normal state and a fault state, obtaining a change amplitude of ksf and a side frequency thereof, wherein k =2n +1, n is a natural number from 1, s is a slip ratio, and f is a base frequency, extracting frequencies which obviously change before and after the fault as characteristic frequencies for diagnosing the fault according to the change rate of the amplitude, and judging whether a rotor winding of the doubly-fed wind driven generator has a three-phase asymmetric load fault;

the step 1 comprises the following steps:

step 1.1: a 5.5kW direct current motor is directly connected with a 5.5kW wound-rotor asynchronous motor to form a double-fed wind generating set, the direct current motor provides external excitation for the asynchronous motor, and the excitation size is controlled and displayed in real time through a double-fed fan excitation control screen; the separately excited direct current motor is directly connected with the output end of the wind turbine characteristic simulation device and receives a variable wind speed regulation signal in real time; one side output end of the asynchronous motor rotor is connected with the rotor side DSP controller to realize synchronous, super-synchronous and sub-synchronous state conversion of the double-fed asynchronous motor, the other side output end is directly connected with a load and a power failure wave recording analysis device, and the power failure wave recording analysis device is finally connected into the power grid side DSP controller to realize electric energy transmission with a power grid; the load is simulated by a bulb;

step 1.2: before the experiment, the performance of a voltage transformer and a current transformer is tested, the voltage transformer is verified through a power frequency power supply, a voltage regulator and a universal meter, the secondary side of the voltage transformer cannot be short-circuited, a resistor is connected in a loop in series, and the voltage regulator and a pincer-shaped ammeter are used for verifying that the current transformer cannot be open-circuited;

step 1.3: when voltage and current signals are collected, in order to avoid the phenomenon that secondary side waveform distortion is caused by the fact that low-frequency signals generate a direct current magnetization effect and exceed the frequency transmission range of a power frequency transformer, a Hall current sensor with the model of HZIA-C06 is selected;

step 1.4: in order to avoid the phenomenon of zero drift caused by the fact that the zero line is not connected on the side of the rotor, any phase is in short circuit with the zero line, and then the voltage value or the current value between any two phases is measured to be used as a correction value to carry out zero drift correction on the measured signal;

the step 2 comprises the following steps:

step 2.1: the wind speed is changed by adjusting the rotating speed of the motor through a wind turbine characteristic simulation device, so that the slip ratio s is changed to set the doubly-fed wind driven generator to respectively operate in three different operation states, including a synchronous state, a subsynchronous state and a supersynchronous state;

step 2.2: selecting any one of the running states of the doubly-fed wind driven generator, when the doubly-fed wind driven generator works in a normal state, acquiring a rotor current or voltage measured value of a measuring point between the asynchronous motor and the converter through a current or voltage transformer, and respectively sending signal waveforms obtained by subtracting a correction value from the measured value to a wave recorder analysis device; filtering the signal waveform by direct current components and carrying out FFT conversion processing to obtain the normal state amplitudes of four frequencies of sf, 3sf, 2-3sf and 2+3sf respectively;

step 2.3: breaking a certain phase bulb on a simulation system to simulate three-phase asymmetric operation faults of a load, acquiring a rotor current or voltage measured value of a measuring point between an asynchronous motor and a converter again, respectively sending signal waveforms obtained by subtracting a correction value from the measured value to an oscillograph analysis device, and respectively obtaining fault state amplitudes of four frequencies of sf, 3sf, 2-3sf and 2+3sf by filtering and FFT conversion processing of the signal waveforms;

the step 3 comprises the following steps:

step 3.1: respectively obtaining the change values between the normal state amplitude and the fault state amplitude of the frequency of the asynchronous motor and the converter at the measuring points sf, 3sf, 2-3sf and 2+3sf under different operating states of the doubly-fed wind generator, and then respectively dividing the change values by the normal state amplitude to obtain the change rate;

step 3.2: selecting the frequency value with the largest change rate from the sf, 3sf, 2-3sf and 2+3sf as the characteristic frequency for diagnosing the three-phase asymmetric load fault;

the step 3 further comprises:

and taking 3sf as the general state characteristic frequency of the doubly-fed wind generator, taking 2-3sf as the subsynchronous state characteristic frequency of the doubly-fed wind generator, and taking 2+3sf as the supersynchronous state characteristic frequency of the doubly-fed wind generator.

Images