A wind turbine main shaft bearing collision and wear fault identification method, system, medium and equipment

Abstract

The invention discloses a method, system, medium and equipment for identifying the friction fault of the main shaft bearing of a wind turbine generator set. The method comprises the following steps: S1, calculating the fault frequency of each component of the main shaft bearing; S2, analyzing the obtained operation data of the wind turbine generator set and the fault frequency of each component of the main shaft bearing, and then extracting the fault characteristic index to form a historical trend curve; S3, setting the warning level threshold for the main shaft bearing of the wind turbine generator set according to the historical trend curve; the present invention is based on the vibration mechanism of the main shaft bearing of the wind turbine generator set, extracts the fault characteristic index in the vibration signal, has a high recognition accuracy, can put forward targeted operation and maintenance opinions according to the real-time operation status of the unit, reduce ineffective operation and maintenance, and then timely discover the initial fault of the wind turbine generator set, perform differential maintenance, ensure the safe operation of the wind turbine generator set, extend the service life of the wind turbine generator set, and avoid the wind turbine generator set from being damaged more.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine main shaft bearing fault identification, and in particular to a wind turbine main shaft bearing rubbing fault identification method, system, medium and equipment. Background Art

[0002] Wind turbines use wind power to drive the blades to rotate. The blades are connected to the main shaft through the hub, and then connected to the gearbox to increase the speed, driving the generator to convert mechanical energy into electrical energy. In general, the speed of the main bearing is low, usually less than 18RPM, and the fault characteristics are easily affected by the vibration noise of other parts in the cabin and environmental noise. Therefore, the fault diagnosis of the main bearing is also a difficult point in the industry.

[0003] At present, the methods for monitoring the operating status and identifying faults of spindle bearings are mainly based on vibration signals, and the main analysis methods are:

[0004] 1. Use support vector machine to establish analysis model, and identify main shaft wear fault through the residual between actual value and predicted value of actual output effective power of generator. Under the same input conditions, when main shaft bearing has wear and other faults, the output active power of generator will gradually decrease as the fault gradually worsens.

[0005] 2. Through the adaptive Hilbert-Huang transform (HHT) time-frequency analysis, the characteristic frequency of the bearing fault can be found from the Hilbert spectrum, and then the location and type of the bearing fault can be identified.

[0006] 3. Use variational modes to decompose, evaluate and analyze the vibration signal of the fan main shaft, select the effective modal components containing signal characteristics to reconstruct the signal and extract the eigenvalues, form a modal analysis vector, and train the VPMCD identification model for fault diagnosis.

[0007] 4. The empirical mode decomposition (MEEMD) method is used to obtain the modal components according to their different central angular frequencies to screen out effective Intrinsic Mode Function (IMF) components, and the mean, root mean square value, peak value, kurtosis, pulse index and other characteristic values ​​of the selected components are calculated. Then, fault identification is performed by combining the CPOS-BBO algorithm with the support vector machine.

[0008] The above analysis methods have a certain effect on the fault identification of the main bearing, but the accuracy often cannot meet the needs of actual projects, and cannot accurately describe the severity of the fault, and cannot provide effective guidance for the operation and maintenance of components. In actual projects, after the wind turbine is connected to the grid, the overall vibration of the unit is large, the noise is noisy, the main shaft speed is low, and the fault frequency is small. It is often manifested in the form of a modulated signal in the spectrum and is easily submerged by the chaotic background noise. Summary of the invention

[0009] The first purpose of the present invention is to solve the deficiencies in the prior art and provide a method for identifying the main shaft bearing friction fault of a wind turbine set. The gearbox meshing frequency and its harmonics become carrier signals, and the bearing defect frequency is modulated on both sides thereof. As the damage develops, the amplitude of the sideband gradually increases. The sideband is used as a fault characteristic indicator to form a historical trend curve. At the same time, an early warning level threshold is set to achieve accurate fault identification and precise determination of the fault level.

[0010] A second object of the present invention is to provide a wind turbine main shaft bearing wear fault identification system.

[0011] A third object of the present invention is to provide a non-transitory computer-readable medium.

[0012] A fourth object of the present invention is to provide a computing device.

[0013] The first object of the present invention is achieved by the following technical solution: A method for identifying a wind turbine main shaft bearing friction fault comprises the following steps:

[0014] S1. Obtaining the operating data of the wind turbine and the parameters of the detection equipment, and calculating the failure frequency of each component of the main shaft bearing;

[0015] S2. Analyze the acquired operation data of the wind turbine and the fault frequencies of various components of the main shaft bearing, and then extract fault characteristic indicators to form a historical trend curve;

[0016] S3. Set a warning level threshold for the main shaft bearing of the wind turbine generator set according to the historical trend curve, so as to identify the fault of the main shaft bearing and determine the fault level.

[0017] Further, in step S1, the following steps are included:

[0018] S101, obtaining operation data of the wind turbine generator set in service based on an online monitoring system of the wind turbine generator set, wherein the operation data includes a vibration signal of a main shaft bearing of the wind turbine generator set and a real-time rotation speed of a wind turbine impeller or a generator;

[0019] S102, obtaining the model and parameters of the detection equipment, and then calculating the fault frequencies of the outer ring, inner ring, rolling element and outer ring of the main shaft bearing, as follows:

[0020] Outer ring fault frequency BPFO calculation formula:

[0021] Inner ring fault frequency BPFI calculation formula:

[0022] Rolling element failure frequency BSF calculation formula:

[0023] The calculation formula of the cage outer ring fault frequency FTF is:

[0024] Among them, r is the bearing speed, n is the number of balls, d is the rolling element diameter, D is the bearing diameter, and α is the rolling element contact angle.

[0025] Further, in step S2, the following steps are included:

[0026] S201, performing low-pass filtering and FFT transformation on the main shaft bearing vibration signal in the operation data of the wind turbine generator set to obtain a carrier signal of the main shaft bearing in a preset frequency band;

[0027] S202. Modulation sidebands with fault frequencies of various components of the spindle bearing as intervals exist at the center frequency and harmonic frequencies of the carrier signal, as follows:

[0028] Calculate the gearbox meshing frequency f m , the calculation formula is:

[0029] f m =(f h +f p )×z h =(f s -f p )×z s

[0030] f p =f b

[0031] Among them, f h is the rotational frequency of the inner gear ring, f p is the planet carrier rotation frequency, z h is the number of teeth on the inner ring gear, f is the rotation frequency of the sun gear, z s is the number of sun gear teeth, f b The main axis rotation frequency;

[0032] When the center frequency f c =f m , carrier signal A m It is expressed as:

[0033] A m =A(t)cos(2πf c t);

[0034] When the signal A(t) is replaced by the low-frequency signal l t Modulation: A(t) = A0[1+l t ];

[0035] Where A0 is the signal amplitude after the carrier signal is modulated, and t is the time;

[0036] If the frequency of the modulated low-frequency signal is f l , then the carrier signal A m It is expressed as:

[0037] A m =A0[1+βcos(2πf l t)]cos(2πf c t);

[0038] Where β is the modulation index;

[0039] S203: Extract fault characteristic indicators from the modulated carrier signal to form a historical trend curve.

[0040] Further, in step S2, the fault characteristic indicators include the energy amplitude of the fault frequency of the outer ring of the main shaft bearing, the energy amplitude of the fault frequency of the inner ring of the main shaft bearing, the energy amplitude of twice the rolling element fault frequency, and the effective value of vibration.

[0041] Further, in step S201, the preset frequency band is 0-200 Hz.

[0042] The second object of the present invention is achieved by the following technical solution: A wind turbine main shaft bearing collision and wear fault identification system, comprising:

[0043] The data reading module is used to obtain the operating data of the wind turbine and detect the equipment parameters;

[0044] A fault frequency calculation module is used to calculate the fault frequency of each component of the spindle bearing;

[0045] A fault characteristic index extraction module extracts fault characteristic indexes according to the modulated carrier signal and the fault frequencies of various components of the spindle bearing to form a historical trend curve;

[0046] The warning level threshold module sets the warning level threshold according to the historical trend curve, so as to identify the faults of the spindle bearings and determine the fault level.

[0047] The third object of the present invention is achieved through the following technical solution: a non-temporary computer-readable medium storing instructions, which, when executed by a processor, executes the steps of the above-mentioned wind turbine main shaft bearing friction fault identification method.

[0048] The fourth purpose of the present invention is achieved through the following technical solution: a computing device, comprising a processor and a memory for storing programs executable by the processor, when the processor executes the program stored in the memory, the above-mentioned wind turbine main shaft bearing friction fault identification method is implemented.

[0049] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0050] 1. Based on the vibration mechanism of the main shaft bearing of the wind turbine, the fault characteristic indicators in the vibration signal are extracted to achieve more accurate diagnosis effect;

[0051] 2. The recognition accuracy is high, and it can put forward targeted operation and maintenance suggestions based on the real-time operation status of the unit to reduce ineffective operation and maintenance;

[0052] 3. Timely discover the initial faults of wind turbines and carry out differentiated maintenance to ensure the safe operation of wind turbines, extend the service life of wind turbines and avoid further damage to wind turbines;

[0053] 4. The method is highly compatible and can be used for various key components of wind turbines of different models. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of the present invention.

[0055] Figure 2 This is the time domain waveform of the complete vibration signal of the faulty main shaft bearing of a wind turbine.

[0056] Figure 3 The figure is the time domain waveform of the filtered faulty spindle bearing vibration signal in the 0-200Hz frequency band.

[0057] Figure 4 The vibration spectrum of the spindle bearing is obtained by performing FFT transformation on the filtered faulty spindle bearing vibration signal.

[0058] Figure 5 It is the historical trend curve of each fault characteristic indicator.

[0059] Figure 6 This is one of the photos showing the repair of a faulty component of a wind turbine main shaft bearing.

[0060] Figure 7 This is the second photo of the repair of the faulty parts of the wind turbine main shaft bearing.

[0061] Figure 8 This is the third photo of the inspection and repair of the faulty components of the wind turbine main shaft bearing.

[0062] Fig. 9 This is the fourth photo of the repair of the faulty components of the main shaft bearing of a wind turbine. DETAILED DESCRIPTION

[0063] The present invention will be further described below in conjunction with specific embodiments.

[0064] Example 1

[0065] See also Figure 1 As shown, the wind turbine main shaft bearing friction fault identification method provided by this embodiment is characterized by comprising the following steps:

[0066] S1. Obtaining the operating data of the wind turbine and the parameters of the detection equipment, and calculating the failure frequency of each component of the main shaft bearing, including the following steps:

[0067] S101, obtaining operation data of the wind turbine generator set in service based on an online monitoring system of the wind turbine generator set, wherein the operation data includes a vibration signal of a main shaft bearing of the wind turbine generator set and a real-time rotation speed of a wind turbine impeller or a generator;

[0068] S102, obtaining the model and parameters of the detection equipment, and then calculating the fault frequencies of the outer ring, inner ring, rolling element and outer ring of the main shaft bearing, as follows:

[0069] Outer ring fault frequency BPFO calculation formula:

[0070] Inner ring fault frequency BPFI calculation formula:

[0071] Rolling element failure frequency BSF calculation formula:

[0072] The calculation formula of the cage outer ring fault frequency FTF is:

[0073] Among them, r is the bearing speed, n is the number of balls, d is the rolling element diameter, D is the bearing diameter, and α is the rolling element contact angle.

[0074] S2. Analyze the acquired operation data of the wind turbine and the fault frequencies of various components of the main shaft bearing, and then extract fault characteristic indicators to form a historical trend curve, including the following steps:

[0075] S201, performing low-pass filtering and FFT transformation on the vibration signal of the main shaft bearing of the wind turbine generator set to obtain a carrier signal of the main shaft bearing in the frequency band of 0-200 Hz;

[0076] S202. Modulation sidebands exist at the center frequency of the carrier signal and its harmonic frequencies, with the fault frequencies of the components of the spindle bearing as intervals, as follows:

[0077] Calculate the gearbox meshing frequency f m , the calculation formula is:

[0078] f m =(f h +f p )×z h =(f s -fp )×z s

[0079] f p =f b

[0080] Among them, f h is the rotational frequency of the inner gear ring, f p is the planet carrier rotation frequency, z h is the number of teeth on the inner ring gear, f is the rotation frequency of the sun gear, z s is the number of sun gear teeth, f b The main axis rotation frequency;

[0081] When the center frequency f c =f m , carrier signal A m It is expressed as:

[0082] A m =A(t)cos(2πf c t);

[0083] When the signal A(t) is replaced by the low-frequency signal l t Modulation: A(t) = A0[1+l t ];

[0084] Where A0 is the signal amplitude after the carrier signal is modulated, and t is the time;

[0085] If the frequency of the modulated low-frequency signal is f l , then the carrier signal A m It is expressed as:

[0086] A m =A0[1+βcos(2πf l t)]cos(2πf c t);

[0087] Where β is the modulation index;

[0088] S203, extracting fault characteristic indicators from the modulated carrier signal to form a historical trend curve, wherein the fault characteristic indicators include the energy amplitude of the fault frequency of the outer ring of the main shaft bearing, the energy amplitude of the fault frequency of the inner ring of the main shaft bearing, the energy amplitude of twice the rolling element fault frequency, and the effective value of vibration.

[0089] S3. Set a warning level threshold for the main shaft bearing of the wind turbine generator set according to the historical trend curve, so as to identify the fault of the main shaft bearing and determine the fault level.

[0090] See also Figure 2As shown in Figure 1, it is a time domain waveform of the complete vibration signal of the main shaft bearing of a wind turbine fault at a certain moment. It can be seen that the characteristic signal of the bearing fault is completely submerged in the background noise; see Figure 3 As shown in FIG. 1 , the complete vibration signal of the above faulty spindle bearing is low-pass filtered to obtain the signal component in the 0-200 Hz frequency band. It can be seen that there is an obvious impact, and the impact interval is the fault frequency of the spindle bearing; see Figure 4 As shown in FIG. 1 , the vibration spectrum of the spindle bearing obtained by performing FFT transformation on the filtered faulty spindle bearing vibration signal shows that there are obvious modulation sidebands at intervals of the spindle bearing fault frequency around its center frequency and its resonance; see Figure 5 As shown in the figure, historical data is taken for analysis. The operation data of the unit has been collected in real time since February 2020. The energy amplitude of the modulation signal 5 times the center frequency on both sides of the vibration spectrum of the main bearing of the unit is extracted, that is, the energy amplitude of the fault frequency of the outer ring of the main shaft bearing (Outer Meshing), the energy amplitude of the fault frequency of the inner ring of the main shaft bearing (Inner Meshing), and the energy amplitude of the 2 times rolling element fault frequency (Roller Meshing) and vibration effective value (RMS), the trend changes within one year are plotted; it can be seen that with the increase of time, the RMS index change trend of the wind turbine is not obvious, while the energy of the modulation signal shows a significant growth trend with the gradual deterioration of the fault; from early March 2020 to November 2020, the index trend of the modulation signal is flat, the main bearing is in an early wear state, and the index amplitude is 6. At this time, the bearing grease should be sufficient and well lubricated; from early November 2020 to April 2021, the index change of the modulation signal increased dramatically, and the main bearing was in a serious wear state with an index amplitude of 8. At this time, on the basis of ensuring good lubrication of the unit, the number of regular inspections of the unit should be increased, and the operating status of the unit should be continuously observed; from early April 2021 to mid-July 2021, the index growth trend slowed down, the main shaft was in a serious deterioration stage, and when the index amplitude was large, it reached 10. In order to ensure the continuous and safe operation of the unit, it must be shut down within two weeks and the main bearing must be replaced; Figures 6 to 9 These are photos of the repair of a faulty main shaft bearing component of a wind turbine. The inner ring, outer ring, and ball bearings all have varying degrees of pitting, the grease has turned gray, and there is a small amount of metal powder.

[0091] Example 2

[0092] This embodiment discloses a wind turbine main shaft bearing collision and wear fault identification system, including:

[0093] The data reading module is used to obtain the operating data of the wind turbine and detect the equipment parameters;

[0094] A fault frequency calculation module is used to calculate the fault frequency of each component of the spindle bearing;

[0095] A fault characteristic index extraction module extracts fault characteristic indexes according to the modulated carrier signal and the fault frequencies of various components of the spindle bearing to form a historical trend curve;

[0096] The warning level threshold module sets the warning level threshold according to the historical trend curve, so as to identify the faults of the spindle bearings and determine the fault level.

[0097] Example 3

[0098] This embodiment discloses a non-transitory computer-readable medium storing instructions. When the instructions are executed by a processor, the steps of the method for identifying abrasion faults of a main shaft bearing of a wind turbine generator set according to Embodiment 1 are executed.

[0099] The non-temporary computer-readable medium in this embodiment may be a disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), a USB flash drive, a mobile hard disk, or other media.

[0100] Example 4

[0101] This embodiment discloses a computing device, including a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the method for identifying the main shaft bearing friction fault of a wind turbine set described in Example 1 is implemented.

[0102] The computing device described in this embodiment may be an embedded host, a desktop computer, a laptop computer, a smart phone, a PDA handheld terminal, a tablet computer, a programmable logic controller (PLC), or other terminal devices with a processor function.

[0103] The embodiments described above are only preferred embodiments of the present invention and are not intended to limit the scope of implementation of the present invention. Therefore, all changes made according to the shape and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying a wind turbine main shaft bearing friction fault, characterized in that: The following steps are involved: S1. Obtaining the operating data of the wind turbine and the parameters of the detection equipment, and calculating the failure frequency of each component of the main shaft bearing, including the following steps: S101, obtaining operation data of the wind turbine generator set in service based on an online monitoring system of the wind turbine generator set, wherein the operation data includes a vibration signal of a main shaft bearing of the wind turbine generator set and a real-time rotation speed of a wind turbine impeller or a generator; S102, obtaining the model and parameters of the detection equipment, and then calculating the fault frequencies of the outer ring, inner ring, rolling element and outer ring of the main shaft bearing, as follows: Outer ring fault frequency BPFO calculation formula: Inner ring fault frequency BPFI calculation formula: Rolling element failure frequency BSF calculation formula: The calculation formula of the cage outer ring fault frequency FTF is: Where r is the bearing speed, n is the number of balls, d is the rolling element diameter, D is the bearing diameter, and α is the rolling element contact angle; S2. Analyze the acquired operation data of the wind turbine and the fault frequencies of various components of the main shaft bearing, and then extract fault characteristic indicators to form a historical trend curve, including the following steps: S201, performing low-pass filtering and FFT transformation on the main shaft bearing vibration signal in the operation data of the wind turbine generator set to obtain a carrier signal of the main shaft bearing in a preset frequency band; S202. Modulation sidebands exist at the center frequency of the carrier signal and its harmonic frequencies, with the fault frequencies of the components of the spindle bearing as intervals, as follows: Calculate the gearbox meshing frequency f m , the calculation formula is: f m =(f h +f p )×z h =(f s -f p )×z s f p =f b Among them, f h is the rotational frequency of the inner gear ring, f p is the planet carrier rotation frequency, z h is the number of teeth on the inner ring gear, f s is the sun wheel rotation frequency, z s is the number of sun gear teeth, f b The main axis rotation frequency; When the center frequency f c =f m , carrier signal A m It is expressed as: A m =A(t)cos(2πf c t); When the signal A(t) is replaced by the low-frequency signal l t Modulation: A(t) = A0[1+l t ]; Where A0 is the signal amplitude after the carrier signal is modulated, and t is the time; If the frequency of the modulated low-frequency signal is f l , then the carrier signal A m It is expressed as: A m =A0[1+βcos(2πf l t)]cos(2πf c t); Where β is the modulation index; S203, extracting fault characteristic indicators from the modulated carrier signal to form a historical trend curve; The fault characteristic indicators include the energy amplitude of the fault frequency of the outer ring of the main shaft bearing, the energy amplitude of the fault frequency of the inner ring of the main shaft bearing, the energy amplitude of twice the rolling element fault frequency and the effective value of vibration; S3. Set a warning level threshold for the main shaft bearing of the wind turbine generator set according to the historical trend curve, so as to identify the fault of the main shaft bearing and determine the fault level.

2. A method for identifying a wind turbine main shaft bearing friction fault according to claim 1, characterized in that: In step S201, the preset frequency band is 0-200 Hz.

3. A wind turbine main shaft bearing friction fault identification system, characterized in that: The system is used to implement the wind turbine main shaft bearing collision and wear fault identification method according to any one of claims 1 to 2, comprising: The data reading module is used to obtain the operating data of the wind turbine and detect the equipment parameters; A fault frequency calculation module is used to calculate the fault frequency of each component of the spindle bearing; A fault characteristic index extraction module extracts fault characteristic indexes according to the modulated carrier signal and the fault frequencies of various components of the spindle bearing to form a historical trend curve; The warning level threshold module sets the warning level threshold according to the historical trend curve, so as to identify the faults of the spindle bearings and determine the fault level.

4. A non-transitory computer-readable medium storing instructions, characterized in that: When the instruction is executed by the processor, the steps of the method for identifying the main shaft bearing friction fault of a wind turbine set according to any one of claims 1 to 2 are executed.

5. A computing device comprising a processor and a memory for storing a program executable by the processor, characterized in that: When the processor executes the program stored in the memory, the method for identifying the main shaft bearing friction fault of a wind turbine set as described in any one of claims 1 to 2 is implemented.

Images