System analysis method for vibration and abnormal sound problems of wind turbine generator

Abstract

The invention belongs to the technical field of data processing, and provides a system analysis method for vibration and abnormal sound problems of a wind turbine generator, which comprises the following steps: determining fault basic information of the wind turbine generator, and obtaining multi-dimensional measured data of the wind turbine generator; on the basis of the fault basic information and the multi-dimensional actual measurement data, online vibration data analysis, unit operation data analysis and maintenance record analysis are conducted on the wind turbine unit, and a preliminary analysis result is obtained; on the basis of the preliminary analysis result, layered troubleshooting is conducted on the complete machine mechanical part of the wind turbine generator, and a first troubleshooting result is obtained; on the basis of the preliminary analysis result, impeller pneumatic unbalance special checking is conducted on the wind turbine generator, and a second checking result is obtained; and according to the first troubleshooting result and the second troubleshooting result, obtaining a system analysis result of the vibration and abnormal sound problems of the wind turbine generator. According to the scheme, the operation stability and reliability of the wind turbine generator are remarkably improved, and efficient and comprehensive troubleshooting of vibration and abnormal sound problems of the wind turbine generator is achieved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a system analysis method for problems such as vibration and abnormal noise in wind turbine generators. Background Technology

[0002] As a core piece of equipment in the clean energy sector, the operational stability of wind turbines directly affects power generation stability and equipment safety. In actual operation and maintenance, problems such as turbine vibration, nacelle swaying, and abnormal gearbox noise frequently occur. If these problems are not investigated and resolved in a timely manner, they can easily lead to escalation of the fault, causing unplanned shutdowns, affecting power generation reliability, and even threatening the structural safety of the equipment.

[0003] Currently, most wind turbine units are equipped with various monitoring devices such as nacelle vibration sensors and online vibration monitoring devices for the drive train, which can collect real-time data on the unit's operating status. However, existing monitoring systems have significant technical limitations. On the one hand, they can only identify characteristic vibration frequencies caused by component damage or failure, and cannot effectively identify vibration and abnormal noise problems caused by non-damage factors such as abnormal stress or assembly deviations. Furthermore, SCADA (Supervisory Control and Data Acquisition) operating parameters often do not provide abnormal feedback, and alarms are only triggered when obvious component failures occur, making it difficult to achieve early warning and accurate tracing of potential hazards.

[0004] On the other hand, in terms of fault analysis, a single-component targeted troubleshooting approach is typically adopted. Technicians often focus only on local components exhibiting signs of failure, lacking an effective method for coherently analyzing all systems of the entire unit and historical maintenance data. This makes it impossible to distinguish between the primary and secondary causes of the fault, leading to recurring problems after multiple repairs and seriously affecting the stable operation of the unit. Furthermore, existing analyses often concentrate on conventional components such as blade appearance, main shaft, and gearbox, neglecting the crucial influencing factor of impeller aerodynamic imbalance, which is in direct contact with external variable loads. This results in analysis results that are difficult to adapt to the needs of complex scenarios.

[0005] This shows that traditional solutions for analyzing vibration and abnormal noise problems in wind turbines are not accurate or reliable enough, and are not efficient or comprehensive enough. Summary of the Invention

[0006] This invention provides a systematic analysis method for wind turbine vibration and abnormal noise problems, which solves the shortcomings of traditional wind turbine vibration and abnormal noise analysis schemes, such as insufficient accuracy and reliability, as well as inefficiency and lack of comprehensiveness.

[0007] This invention provides a systematic analysis method for vibration and abnormal noise problems in wind turbine generators, including: Determine the basic fault information of the wind turbine and obtain multi-dimensional measured data of the wind turbine; Based on the fault information and the multidimensional measured data, online vibration data analysis, unit operation data analysis and maintenance record analysis were performed on the wind turbine to obtain preliminary analysis results. Based on the preliminary analysis results, the mechanical components of the wind turbine were investigated in layers to obtain the first investigation results; Based on the preliminary analysis results, a special investigation was conducted on the aerodynamic imbalance of the wind turbine rotor, and the second investigation results were obtained. Based on the first and second investigation results, a system analysis of wind turbine vibration and abnormal noise issues was obtained.

[0008] According to the system analysis method for wind turbine vibration and abnormal noise problems provided by the present invention, the basic fault information of the wind turbine is determined, including: Determine the occurrence mode, location, and conditions of vibration and abnormal noise problems in wind turbine units; The occurrence mode, location, and conditions are used as the basic fault information for wind turbine units.

[0009] According to the system analysis method for wind turbine vibration and abnormal noise problems provided by the present invention, the multi-dimensional measured data includes: the original vibration data during the period when the wind turbine failure occurred; Online vibration data analysis of the wind turbine generator includes: The original vibration data were subjected to time-domain feature analysis and frequency-domain feature analysis respectively to obtain online vibration characteristic parameters; The online vibration characteristic parameters are compared with the reference vibration characteristic parameters obtained in advance during historical normal operation periods to obtain the first comparison result; Based on the first comparison result and combined with the basic fault information, the damaged component and the degree of damage to the damaged component are determined, and the online vibration data analysis results are obtained.

[0010] According to the system analysis method for wind turbine vibration and abnormal noise problems provided by the present invention, the multi-dimensional measured data includes: second-level operating data of the wind turbine within a set time period before and after the fault. The wind turbine generator set is subjected to unit operation data analysis, including: The second-level operating data of the wind turbine unit within a set time period before and after the fault is compared with the normal operating data of other units in the same period obtained in advance to obtain a second comparison result; Based on the second comparison result, the operational changes of the wind turbine unit before and after the fault and other normal units in the same time period are analyzed. Combined with the basic fault information, the abnormal points in the wind turbine unit are preliminarily determined, and the unit operation data analysis results are obtained.

[0011] According to the system analysis method for vibration and abnormal noise problems of wind turbine units provided by the present invention, the second-level operating data includes: real-time wind speed, operating speed, active power, main shaft gearbox temperature, generator bearing temperature, oil temperature, pitch angle, pitch speed, and yaw system pressure.

[0012] According to the system analysis method for vibration and abnormal noise problems of wind turbine units provided by the present invention, the multi-dimensional measured data includes: historical maintenance and repair records of problematic components in the wind turbine unit; The maintenance and repair records of the wind turbine units were analyzed, including: Based on the historical maintenance records, the wind turbine units were located and the locations of frequently faulty components were identified to determine the areas where historical faults were concentrated. Based on the aforementioned basic fault information, the analysis of influencing factors, diagnosis of the rationality of solutions, quality inspection and acceptance verification, and verification of key assembly dimensions are conducted on the areas where historical faults are concentrated, resulting in the analysis results of maintenance records.

[0013] According to the system analysis method for vibration and abnormal noise problems of wind turbine units provided by the present invention, based on the preliminary analysis results, a layered investigation of the mechanical components of the wind turbine unit is performed to obtain a first investigation result, including: Based on the preliminary analysis results, the mechanical components of the wind turbine were inspected at the foundation layer, tower layer, and nacelle interior and related components, respectively, and the inspection sub-results for each layer were obtained. The investigation results corresponding to each layer are summarized to obtain the first investigation result.

[0014] According to the system analysis method for vibration and abnormal noise problems of wind turbine units provided by the present invention, the mechanical components of the wind turbine unit are inspected, including the internal structure of the nacelle and related components. The troubleshooting should proceed in order from the generator to the impeller, starting with the closest to the furthest point. Specifically, this includes: The generator, coupling, gearbox, main shaft, nacelle, yaw system, pitch system, and impeller were inspected in sequence.

[0015] According to the system analysis method for wind turbine vibration and abnormal noise problems provided by the present invention, based on the preliminary analysis results, a special investigation of impeller aerodynamic imbalance is conducted on the wind turbine to obtain a second investigation result, including: Obtain measured images of the impeller; Based on the measured images of the impeller, the actual distance from the blade tip to the tower is determined, a trend diagram of the net clearance change of the three blades is generated, and the degree of imbalance is determined according to the vibration velocity classification. Based on the aforementioned imbalance and the preliminary analysis results, the aerodynamic imbalance influencing factors of the impeller were investigated, and the second investigation results were obtained.

[0016] According to the system analysis method for wind turbine vibration and abnormal noise problems provided by the present invention, based on the unbalance degree and combined with the preliminary analysis results, the aerodynamic unbalance influencing factors of the impeller are investigated to obtain a second investigation result, including: Based on the aforementioned imbalance and the preliminary analysis results, the blade mass moment deviation was checked, the blade electrical zero position was calibrated, the deviation between the zero-degree scale and the positioning block was detected, the yaw positioning accuracy was investigated, and the pitch system coordination was tested to obtain the second investigation results.

[0017] The system analysis method for vibration and abnormal noise problems in wind turbines provided by this invention first clarifies the basic fault information and integrates multi-dimensional measured data. Then, it uses multi-dimensional data analysis to form a precise guide. Subsequently, it conducts a layered investigation of the entire machine's mechanical components and a specific diagnosis of impeller aerodynamic imbalance in parallel. Finally, it integrates the results of the dual investigations to form a closed-loop analysis conclusion. It not only achieves comprehensive coverage of all system components of the entire machine, such as the foundation, tower, and transmission chain, and accurately distinguishes between primary and secondary causes of the fault, but also specifically solves the diagnostic challenge of impeller aerodynamic imbalance, a key hidden influencing factor. This ensures the comprehensiveness and accuracy of fault analysis, effectively avoids multiple ineffective repairs, and can prevent the unit from shutting down and structural damage due to the expansion of the fault in advance. It significantly improves the operational stability and reliability of wind turbines and achieves efficient and comprehensive investigation of vibration and abnormal noise problems in wind turbines. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the system analysis method for wind turbine vibration and abnormal noise problems provided in this embodiment of the invention. Figure 2 This is a diagram showing the causes and effects of cabin vibration failure. Figure 3 This is a fault cause correlation diagram for abnormal noises from the generator unit; Figure 4 This is a schematic diagram illustrating the analysis of transmission chain slippage and tilting; Figure 5 This is the blade clearance change curve under a mild aerodynamic imbalance scenario; Figure 6 This is the blade clearance change curve under a severe aerodynamic imbalance scenario; Figure 7 This is a detailed flowchart illustrating the analysis process of a Python program; Figure 8 This is a schematic diagram showing the measurement relationship between the zero mark and the limit stop at the blade root flange and the parting line; Figure 9 This is a schematic diagram of the calibration reference for the angle and arc length between the zero mark and the first flange bolt; Figure 10 This is a schematic diagram of the actual on-site measurement of the hydraulic pitch angle. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] The following is combined Figures 1 to 10 This invention describes the detailed scheme of the system analysis method for wind turbine vibration and abnormal noise problems provided in the embodiments of the present invention.

[0022] like Figure 1 As shown in the figure, the system analysis method for wind turbine vibration and abnormal noise problems provided in this embodiment of the invention mainly includes the following steps: Step 110: Determine the basic fault information of the wind turbine and obtain multi-dimensional measured data of the wind turbine.

[0023] Understandably, defining the basic fault information can provide boundary conditions for subsequent analysis, and acquiring multi-dimensional measured data can provide effective data support for subsequent system analysis.

[0024] Step 120: Based on the basic fault information and multi-dimensional measured data, online vibration data analysis, unit operation data analysis, and maintenance record analysis are performed on the wind turbine to obtain preliminary analysis results.

[0025] In this embodiment, through multi-dimensional data analysis, abnormal factors can be initially identified, providing a preliminary basis for subsequent layered investigation of the mechanical components of the whole machine.

[0026] Step 130: Based on the preliminary analysis results, conduct a layered inspection of the mechanical components of the wind turbine to obtain the first inspection results.

[0027] Understandably, by conducting a layered inspection of the mechanical components of the entire machine, the data conclusions can be preliminarily verified at the on-site data level, providing some basis for the conclusions of the analysis.

[0028] Step 140: Based on the preliminary analysis results, a special investigation was conducted on the aerodynamic imbalance of the wind turbine rotor to obtain the second investigation results.

[0029] In this embodiment, the special investigation process for impeller aerodynamic imbalance can focus on core influencing factors and delve deeper to locate the root cause, thereby enabling a deeper understanding of vibration and abnormal noise issues and further improving the accuracy of system analysis results.

[0030] Step 150: Based on the results of the first and second investigations, obtain the system analysis results of the vibration and abnormal noise problems of the wind turbine.

[0031] The solution provided by this invention breaks through the blind spot of thinking that wind turbine vibration and abnormal noise fault analysis is limited to a single faulty component. It designs a set of influencing factor analysis schemes based on all related mechanical components of the whole machine, which can comprehensively cover all system components such as foundation, tower, nacelle, and transmission chain. It can accurately locate the root cause of the fault and improve the comprehensiveness, accuracy and effectiveness of the analysis of difficult fault causes.

[0032] In one embodiment, determining the basic fault information of the wind turbine specifically includes: First, determine the form, location, and conditions under which vibration and abnormal noise problems occur in wind turbine units.

[0033] Then, the occurrence mode, location, and conditions are used as the basic fault information for the wind turbine.

[0034] In practical applications, it is necessary to determine the form of nacelle vibration, such as up-and-down nodding, left-and-right swaying, or overall nacelle vibration, and to determine the location of the vibration, clarifying whether the vibration occurs in the transmission chain components, the nacelle, or the tower, in order to initially determine the approximate direction and scope of the analysis.

[0035] Next, it's crucial to understand the conditions under which nacelle vibration occurs. For example, was the wind turbine's location during construction particularly sensitive to external wind conditions? It's also essential to identify basic information such as ambient temperature, wind speed range, rotor input speed, generator speed, active power, and yaw / pitch angle at the time of nacelle vibration to determine whether the vibration occurs under specific conditions or is a persistent issue, thus allowing for the analysis of related factors. Simultaneously, the natural frequencies of the blades and tower need to be checked to rule out resonance, and it's necessary to examine whether any changes to the control program and system protection settings have caused any impact.

[0036] In one embodiment, the multidimensional measured data specifically includes: the raw vibration data during the period when the wind turbine failure occurred.

[0037] In practical applications, raw vibration data for the time period during which the fault occurred can be obtained, focusing on the vibration monitoring channel data corresponding to the spindle, gearbox, and generator to ensure that there is no data loss or distortion.

[0038] Even better, filtering algorithms can be used to remove noise signals caused by environmental interference such as wind noise and electromagnetic interference, perform stationarity tests on the data, and perform detrending processing if a trend term exists to ensure that the analysis object is a pure vibration signal.

[0039] Furthermore, online vibration data analysis of the wind turbine generators is conducted, specifically including: First, time-domain and frequency-domain feature analyses were performed on the raw vibration data to obtain online vibration characteristic parameters.

[0040] In the time-domain analysis stage, characteristic indicators such as peak value, effective value, peak factor, and kurtosis of the vibration signal in the original vibration data can be calculated and compared with the factory standard value or industry specification value of the equipment: if the effective value exceeds the upper limit or the kurtosis value increases significantly, it can indicate that there may be local damage such as bearing pitting or gear cracks.

[0041] In the frequency domain analysis phase, the time-domain signal can be converted into a frequency-domain spectrum using a Fast Fourier Transform (FFT) to identify characteristic frequencies. Specifically, for spindle bearings, the characteristic frequencies of the inner ring, outer ring, rolling elements, and cage can be calculated. If a peak value of the corresponding frequency appears in the spectrum and is accompanied by an overtone, the bearing is considered damaged. For gearboxes, the gear meshing frequency and harmonic frequencies can be extracted. If abnormal peak values ​​of the meshing frequency, obvious sidebands, or abundant harmonic frequencies are observed, the gears are considered to have wear, broken teeth, or other damage. Simultaneously, the bearing characteristic frequencies are analyzed to troubleshoot bearing faults within the gearbox. For generators, the focus is on analyzing the bearing characteristic frequencies and rotor imbalance frequencies. If the peak value of the bearing characteristic frequency exceeds the standard, the bearing is considered damaged. If the fundamental frequency peak is significant and accompanied by an overtone, rotor faults need to be investigated in conjunction with dynamic balancing data.

[0042] Then, the online vibration characteristic parameters are compared with the reference vibration characteristic parameters of the historical normal operation period obtained in advance to obtain the first comparison result.

[0043] In practical applications, the online vibration characteristic parameters can be compared with the reference vibration characteristic parameters during historical normal operation periods. If the difference rate exceeds 30% and the characteristic frequency persists, further confirmation of damage is required.

[0044] Finally, based on the first comparison results and combined with the basic fault information, the damaged components and the degree of damage to the damaged components are determined, and the online vibration data analysis results are obtained.

[0045] In this embodiment, the degree of damage can be determined based on the ratio of the peak value of the characteristic frequency to the standard value and the richness of the harmonics. Specifically, a peak value exceeding the standard value by 1-2 times is considered mild damage, 2-3 times is considered moderate damage, and exceeding 3 times or the presence of multiple harmonics is considered severe damage, thus providing a basis for subsequent maintenance priorities.

[0046] In this embodiment, online vibration data analysis can determine whether there is damage to key components such as the main shaft, gearbox, and generator bearings or gears, thus achieving online vibration data diagnosis.

[0047] In one embodiment, the multidimensional measured data also includes: second-level operating data within a set time period before and after the wind turbine failure.

[0048] Specifically, the second-level operating data includes: real-time wind speed, operating speed, active power, main shaft gearbox temperature, generator bearing temperature, oil temperature, pitch angle, pitch speed, and yaw system pressure.

[0049] Furthermore, the wind turbine's operating data is analyzed, specifically including: First, the second-level operating data of the wind turbine unit within a set time period before and after the failure is compared with the normal operating data of other units during the same period obtained in advance, and a second comparison result is obtained.

[0050] Then, based on the second comparison results, the operational changes of the wind turbine unit before and after the fault and other normal units in the same time period are analyzed. Combined with the basic fault information, the abnormal points in the wind turbine unit are preliminarily determined, and the unit operation data analysis results are obtained.

[0051] In this embodiment, by analyzing the unit's operating data, we can check whether the temperatures of the main shaft, gearbox, and generator bearings are normal, whether the pitch angles of the three blades are consistent or within the normal deviation range, whether the yaw pitch rate and frequency are too large, whether the hydraulic system's full and partial unloading pressures meet the requirements, and whether the unit is affected by turbulence. We can also analyze whether there are any abnormalities in the unit's power curve, power generation efficiency, and grid voltage and current. Furthermore, we can compare the second-level operating data of the wind turbine unit within a set time period before and after the fault with the pre-obtained normal operating data of other units during the same period to identify the abnormality.

[0052] In one embodiment, the multidimensional measured data also includes: historical inspection and maintenance records of problematic components in the wind turbine.

[0053] Specifically, historical maintenance records of major components of the generator set, including basic components, tower, transmission chain, yaw system, and pitch system, throughout their entire life cycle can be collected and categorized by fault time, component name, fault location, fault type, maintenance content, and maintenance frequency to form a structured record list.

[0054] Furthermore, the maintenance and repair records of the wind turbine units were analyzed, specifically including: First, based on historical maintenance records, the locations of frequently faulty components of the wind turbine were identified to determine the areas where historical faults were concentrated.

[0055] In practical applications, the number of failures and the frequency of repeated repairs for each component in the structured record list can be counted to identify frequently failing components and their specific locations, thereby clarifying the areas where historical failures are concentrated.

[0056] Then, based on the basic fault information, we conducted an analysis of the influencing factors, a diagnosis of the rationality of the proposed solutions, a quality inspection and acceptance review, and a verification of the dimensions of critical assembly in areas with concentrated historical faults, and obtained the results of the maintenance and repair record analysis.

[0057] Understandably, the analysis of maintenance records mainly traces the types, causes, maintenance content, and frequency of failures that occurred during the life cycle of major components, thereby conducting a preliminary analysis of possible factors that could cause unit failures.

[0058] In the influencing factor analysis stage, by combining failure modes such as wear, cracking, and jamming with the corresponding component's design specifications and operating conditions, potential influencing factors can be preliminarily identified, such as assembly deviations, insufficient lubrication, and material fatigue. In the solution rationality diagnosis stage, the plans for carried out maintenance projects can be reviewed to determine whether the maintenance measures address the root cause of the failure and to assess whether the maintenance process and spare parts used meet the equipment's technical requirements. In the quality inspection and acceptance review stage, the quality inspection report after maintenance can be reviewed to confirm whether the acceptance items cover key performance indicators, whether the inspection data meets standards, and whether there are any unrectified quality hazards. In the critical assembly dimension verification stage, the critical assembly dimensions recorded during maintenance can be extracted and compared with design standard values ​​to determine whether there are dimensional deviations and to analyze whether these deviations lead to abnormal component stress or increased vibration.

[0059] Finally, conclusions can be summarized and doubts marked. The analysis results can be summarized to clarify the correlation between historical faults and current vibrations and abnormal noises, and key quality issues such as dimensional deviations and inadequate maintenance can be marked to provide key directions for subsequent mechanical component inspections.

[0060] In one embodiment, combined with Figure 2 , Figure 3 as well as Figure 4 As shown, based on the preliminary analysis results, a layered inspection of the mechanical components of the wind turbine was conducted, resulting in the first inspection result, which specifically includes: First, based on the preliminary analysis results, the mechanical components of the wind turbine were inspected at the foundation layer, tower layer, and nacelle interior and related components, respectively, and the inspection results for each layer were obtained.

[0061] During the foundation inspection phase, the main focus is on checking for cracks, loosening, subsidence, and water erosion in the wind turbine foundation concrete, as well as identifying issues that could affect the stability of the foundation and cause unit shaking.

[0062] During the tower inspection phase, the focus is on checking for cracks in the tower welds, whether the gaps between the flange connections are uniform, whether the high-strength connecting bolts are broken or loose, and whether the cables swing and impact the fixing devices or the tower, which could cause uneven tower connections and stress, or swaying due to external loads.

[0063] In a specific implementation, the mechanical components of the wind turbine are inspected, including the interior of the nacelle and related parts. The troubleshooting process proceeds in order from the generator to the impeller, starting with the closest components and moving outwards. This process specifically includes: The generator, coupling, gearbox, main shaft, nacelle, yaw system, pitch system, and impeller were inspected in sequence.

[0064] Combination Figure 2 and Figure 3As shown, during generator troubleshooting, it is necessary to check whether the generator bearings are damaged, whether the grease is caked or blackened, whether the slip rings and carbon brushes are abnormally worn, whether the anchor bolts are missing or loose, whether the electrical winding parameters are normal, whether there are any cracks or burns, whether there are any abnormal noises from the bearings and obvious vibrations from the gearbox during operation, and at the same time verify whether the generator alignment record and alignment data are qualified. If they are not qualified, it will cause abnormal meshing of the gearbox gears and bearings, resulting in vibration, and will increase wear and further cause abnormal noises.

[0065] During the coupling inspection process, it is necessary to check whether the coupling and connecting diaphragm are damaged or deformed, and to conduct operational tests to check whether there is any slippage at the diaphragm connection points, and whether there are any abnormal noises or shaking during operation.

[0066] During gearbox inspection, the gears and bearings inside the gearbox can be examined with an endoscope to check for raised spots, fretting corrosion, abnormal wear scratches, dents, deformation, and damage that could cause vibration and abnormal noise. The meshing indentation can be used to determine if there is any uneven load on the shaft system. A feeler gauge can be used to check if the bearing outer ring clearance meets requirements. The gears and bearing rollers can be checked for rotation under no-load conditions to identify any interference factors during operation. Special attention should be paid to checking for damage or aging of the elastic support pads. The elastic support pads should be manually rotated circumferentially to check for obvious looseness, and a feeler gauge should be used to check if the vertical clearance of the elastic support pads is excessive. The distance difference between the left and right support arms and the front and rear bearing seats should be measured separately.

[0067] like Figure 4 As shown, the values ​​of L1, L2, L3, and L4 can be measured and compared. If [L1(L3)-L2(L4)]>5mm or [L5(L7)-L6(L8)]>5mm, the gearbox has a large axial movement. If L1>L3>L4>L2, the gearbox moves backward and tilts to the left. If L3>L1>L2>L4, the gearbox moves backward and tilts to the right. If (L1≈L4)>(L2≈L3), the gearbox tilts to the left. If (L2≈L3)>(L1≈L4), the gearbox tilts to the right. All of these are due to varying degrees of external load fluctuations, large impacts, or abnormal loads on gears and bearings, which affect the vibration of the unit. At the same time, the location of vibration and abnormal noise can be located by running tests, and it can be checked whether the support arm floats up and down significantly during operation. If it floats significantly, the elastic support pad is ineffective or the vertical gap is too large.

[0068] During spindle inspection, check the bearing roller raceways and cage for obvious pitting, peeling, or damage. Inspect the cage for significant curling that could cause excessive clearance between the rollers and cage. Shake the rollers to check for looseness or strain. Check the grease for obvious hardening or blackening, and for excessive metal filings. Measure the clearance between the inner side of the bearing housing's outer stop and the outer ring at eight evenly spaced points around the circumference to ensure uniformity and close contact. Uneven or excessive clearance may indicate spindle movement or tilting. Simultaneously, use a level to measure the spindle elevation angle, which should be around 5°. An angle that is too large or too small indicates spindle tilting or lifting, leading to abnormal bearing stress. Simultaneously measure the distance from the left and right ends of the spindle flange to the frame and compare it with the design dimensions. If the difference is more than 5mm, there is significant movement; if the left and right distance deviation is >3mm, the transmission chain is tilted, which can cause abnormal load on the planetary gears and parallel bearings, resulting in vibration or abnormal noise. Check for loose or broken bearing housing bolts, and for any displacement or misalignment of the bearing housing. Perform a low-speed test on the spindle to determine if there is any significant vibration or abnormal noise.

[0069] During the inspection of the main body of the engine room, you can check for cracks in the welded parts of the front and rear bases of the engine room, loose or missing bolts connecting the upper and lower engine rooms, and loose or wobbling engine room accessories.

[0070] During the yaw system inspection, you can check for severe wear and unevenness of the yaw brake disc or pads, hard particles embedded in the mating surface, measure the change in installation height to determine if the brake disc is deformed, test the yaw process and yaw motor operation for jamming and vibration, abnormal noise from the bearings, impact between the meshing surfaces of the yaw gears, abnormal force caused by asynchronous yaw motors, and whether insufficient hydraulic braking force causes engine compartment slippage, resulting in vibration or abnormal noise.

[0071] During the pitch system and impeller inspection, the blades, hub, bearings, pitch gear ring, reducer output wheel, pitch motor or hydraulic cylinder body, and connecting bolts can be inspected. Visually, the outer surface of the blades can be inspected with binoculars or drones for cracks, wrinkles, lightning strikes, and severe wind erosion. The inner surface can be inspected manually or by robots for damage. The hub surface can be inspected for cracks, and the bearing connecting bolts can be checked for looseness or breakage. The bearings can be inspected for damage, and the zero-degree teeth of the pitch gear ring and reducer output wheel can be checked for obvious wear or wear depth exceeding tolerance, which may cause pitch vibration or abnormal noise.

[0072] For hydraulic pitch control mechanisms, check for oil leaks in the hydraulic cylinders and for loose or damaged mechanical parts. Simultaneously test the opening and closing of the three blades to check the timeliness, smoothness, and accuracy of the rotation angle. Confirm the absence of impacts, vibrations, and delays, and ensure the pitch control process is smooth, without vibration, abnormal noise, or jamming. Simultaneously observe whether the three pitch motors operate normally, checking for abnormal noises from the motor blades and internal mechanical parts, and ensuring the brake pads provide timely and smooth braking. For hydraulic pitch control mechanisms, check for any jamming or abnormal noises in the mechanical parts.

[0073] Then, the investigation results corresponding to each layer are summarized to obtain the first investigation result.

[0074] In one embodiment, based on the preliminary analysis results, a special investigation of rotor aerodynamic imbalance is conducted on the wind turbine, resulting in a second investigation result, which specifically includes: First, obtain measured images of the impeller.

[0075] During the impeller aerodynamic imbalance identification stage, in clear weather with wind speeds greater than 6 m / s and relatively stable wind speeds, record a video of impeller rotation for more than 1 minute from a distance of more than 100 meters from the side of the wind turbine (vertical to the direction of the wind turbine nacelle). The video must always include three blades, the nacelle, and the tower. If due to terrain limitations, the video should at least include the position of each blade at its lowest point and an image of the tower to ensure that the shape of the tower and blades and the changing trend of the distance between the blade tip and the tower can be captured throughout the video analysis. The camera should be kept fixed and without shaking during the recording process.

[0076] Then, based on the measured images of the impeller, the actual distance from the blade tip to the tower is determined, a trend diagram of the net clearance change of the three blades is generated, and the degree of imbalance is determined according to the vibration velocity classification.

[0077] Furthermore, the recorded video is imported into a blade clearance analysis program designed in Python. This program can analyze the video images frame by frame. After the analysis is completed, it will generate a clearance change diagram of the three blades, which can show the process of the actual distance from the blade tip to the lowest point of the tower changing over time.

[0078] like Figure 5 and Figure 6 As shown in the figure, the curves of different colors correspond to the real-time clearance values ​​of the three blades, that is, the actual distance from the blade tip to the tower. The horizontal axis represents time, and the vertical axis represents the actual distance. Figure 5 The curves of the three blades in the middle section have small fluctuations and a consistent overall trend, indicating that the blade clearance changes uniformly, corresponding to a scenario with mild aerodynamic imbalance. Figure 6The significant spike fluctuations in some blades, with a marked difference in curves compared to other blades, indicate an abnormal change in the headroom of these blades, corresponding to a moderate to severe aerodynamic imbalance scenario. Therefore, the degree of impeller imbalance can be determined by observing the headroom changes of each blade and the headroom differences among the three blades.

[0079] Specifically, vibration velocities of around 6.3 mm / s to 8.44 mm / s are generally considered to be slightly unbalanced; around 8.44 mm / s to 14.07 mm / s are considered to be moderately unbalanced; and above 14.07 mm / s are considered to be severely unbalanced.

[0080] Finally, based on the imbalance degree and combined with the preliminary analysis results, the aerodynamic imbalance influencing factors of the impeller were investigated, and the second investigation results were obtained.

[0081] In a specific implementation, based on the degree of imbalance and combined with the preliminary analysis results, the aerodynamic imbalance influencing factors of the impeller are investigated, resulting in a second investigation result, which specifically includes: Based on the imbalance and combined with the preliminary analysis results, the blade mass moment deviation was checked, the blade electrical zero position was calibrated, the deviation between the zero degree scale and the positioning block was detected, the yaw positioning accuracy was investigated, and the pitch system coordination was tested, resulting in the second investigation results.

[0082] In this embodiment, the detailed analysis process of the compiled Python program can be found in [reference needed]. Figure 7 The specific steps are as follows: The first step is program startup and initialization: After the program starts, it first performs module initialization operations, involving the main control module, vision processing module, database module, and visualization module, to prepare for the smooth operation of subsequent functions. The database module is used for database-related operations, the vision processing module is used for handling vision-related tasks, the visualization module is used for visualization-related operations, and the main processing module is used for overall process control.

[0083] The second step is the calibration process: Next, the calibration judgment stage is entered to determine whether forced calibration is required. If forced calibration is not required, the default pixel ratio is used directly; if forced calibration is required, the first-frame calibration process is executed, which involves detecting the tower and blades, calculating the pixel ratio, and then entering the automatic monitoring loop after calibration is completed.

[0084] The third step is the automatic monitoring loop: In the automatic monitoring loop, video frames are first read and their validity is determined. If the frame is invalid, subsequent steps in the process will not proceed; if the frame is valid, tower detection and blade tracking operations will be performed.

[0085] The fourth step is result processing: determining whether the tower and blades were successfully detected. If the detection is unsuccessful, an anomaly is marked and the distance calculation is skipped; if the detection is successful, the actual distance between the blades and the tower is calculated, and this distance data is stored in the database along with historical data.

[0086] Step 5: Visualization and Program Termination: Whether the detection was successful and distance calculation was performed, or the detection failed and anomalies were marked, visualization operations will be performed, drawing dynamic trend charts of the tower and blade trajectories, and performing related visualizations such as FFT (Fast Fourier Transform). Finally, the program will terminate and resources will be released, including closing the video, windows, and database.

[0087] In the process of investigating and analyzing factors affecting impeller aerodynamic imbalance, after determining that there is an imbalance in the impeller using the above methods, the following aspects need to be investigated: On the one hand, investigate the deviation in the mass moment of the three blades caused by poor manufacturing process at the time of manufacture or substandard counterweight after repair. The specific steps are as follows: The first step is to calculate the total mass m´ after the counterweight is applied.

[0088] The second step is to calculate the equivalent center of gravity radius of the blade after the counterweight is applied. If the original center of gravity of the counterweight and the blade are not on the same rotation radius line, remeasure the position of the center of gravity of the blade after the counterweight is applied and calculate the equivalent center of gravity radius L´, which is the distance from the center of gravity of the blade after the counterweight is applied to the reference support point plus the distance from the reference support point to the center of rotation of the hub.

[0089] The third step is to calculate the mass moment M´=m´×L´ after the single blade is counterweighted.

[0090] The fourth step is to calculate the average moment of mass after the three blades are counterweighted.

[0091] The fifth step is to calculate the relative deviation ΔM´ of the mass moment of a single blade after weighting, relative to the average mass moment. If the relative deviation ΔM´ of the mass moment of all blades after weighting is ≤0.1%, the mass moment check is deemed qualified; if the deviation of any blade exceeds the limit, the check is deemed unqualified, and the weighting needs to be redone until it is qualified.

[0092] As can be understood, the mass moment is a physical quantity that describes the magnitude of an object's inertia when it rotates about a certain axis. Its core meaning is the object's ability to resist changes in its rotational state, and it is directly related to the object's mass distribution and the position of the rotation axis.

[0093] On the other hand, check for excessive cumulative errors in the electrical zero-position operation of the blades. Specifically, the control system can be used to control each blade to pitch to 0°, and check whether the "0" mark on the scale attached to the outer root of each blade is aligned with the pointer on the hub. If they are not aligned, record the actual angle value of each blade after zeroing, and compare the angle values ​​of the three blades. Consult the unit manual to determine the deviation range required by the manufacturer. If the deviation exceeds the limit, the blade with the largest zero-angle deviation will be re-zeroed to the required range.

[0094] On the other hand, investigate manufacturing and installation deviations in the blade pitch zero-degree scale or positioning stop. Due to reasons such as manufacturer's processes, management, or quality control, some blades leaving the factory may have incorrect markings on the zero-degree scale or positioning stop. (Refer to...) Figure 8 , Figure 9 as well as Figure 10 As shown, the specific detection methods include the following aspects: First, check the accuracy of the zero-scale marking by verifying the theoretical arc length S1 or chord length L1 of the edge profile at the inner diameter of the blade root flange from the zero-scale line on the zero-scale gauge. (Refer to...) Figure 8 As shown, the difference between the theoretical arc length S2 or chord length L2 between the leading edge parting line and the nearest inner side of the limiting block at the inner diameter of the blade root flange can also be verified. If the difference between the distances of the three blades from the chord length exceeds the design range, it will cause the actual blade angles to be inconsistent, resulting in impeller aerodynamic imbalance, which needs to be adjusted.

[0095] Understandably, the parting line refers to the outline left on the surface of the finished blade by the contact surface of the blade mold (usually the upper and lower molds). Essentially, it is the splicing mark between the upper and lower molds during blade forming.

[0096] Secondly, tooling verification can also be used during the production process. The root verification method involves measuring the arc length or chord length of the first bolt hole at the flange outer diameter from the zero-scale position in the counterclockwise direction (i.e., from the blade root to the blade tip). Figure 9 As shown, the arc length S1 of the flange inner diameter corresponds to angle α.

[0097] Third, for hydraulic pitch control units, such as Figure 10 As shown, first use the TC line special inspection tool template to determine the position of the 25.5° extension line of the blade root. Then, use a long ruler to align with the center line of the blade PIN bolt and measure the distance L between the center line of the blade PIN bolt and the 25.5° extension line. Measure the distance L values ​​for each of the three blades and compare the differences. If there is a blade with a large difference, its actual angle will be inconsistent with the other two blades, resulting in aerodynamic imbalance, which needs to be adjusted.

[0098] It is understandable that the TC line is the reference tangent line on the blade root section (i.e. the end section connected to the hub) along the blade chord direction (i.e. the direction of the line connecting the leading edge to the trailing edge).

[0099] On the other hand, troubleshoot inaccurate yaw positioning and malfunctions in yaw system components. Check whether the wind vane or sensors are faulty, causing inaccurate yaw positioning. Uneven wind force on the wind turbine will result in significant impact and vibration during yaw, which will be transmitted to the nacelle and cause vibration.

[0100] In practical applications, by combining the results of the first and second investigations, and through the logical integration of factor classification, primary and secondary determination, and causal verification, the system analysis results are output. Specifically, the faults of components such as the foundation, tower, and transmission chain in the first investigation results can be classified and summarized with the types and specific causes of impeller aerodynamic imbalance in the second investigation results. Then, based on the correlation between the vibration frequency, abnormal noise characteristics, and various fault factors in the preliminary analysis results, the core primary cause and secondary causes can be determined. Finally, through cross-validation, the root cause, degree of impact, and transmission path of the fault are clarified, forming a complete system analysis result that includes root cause location, damage level, rectification plan, and preventive measures, thereby providing a basis for precise maintenance and operation optimization.

[0101] In summary, this invention details the systematic analysis steps for wind turbine vibration and abnormal noise problems, points out the direct impact of impeller aerodynamic imbalance on turbine vibration and abnormal noise problems, and explains the detailed analysis process of the influencing factors of impeller aerodynamic imbalance. By identifying impeller aerodynamic imbalance problems in a more efficient, convenient, and accurate way, it can effectively avoid turbine tower collapse and damage to major components.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A system analysis method for vibration and abnormal noise problems of wind turbine generators, characterized in that, include: Determine the basic fault information of the wind turbine and obtain multi-dimensional measured data of the wind turbine; Based on the fault information and the multidimensional measured data, online vibration data analysis, unit operation data analysis and maintenance record analysis were performed on the wind turbine to obtain preliminary analysis results. Based on the preliminary analysis results, the mechanical components of the wind turbine were investigated in layers to obtain the first investigation results; Based on the preliminary analysis results, a special investigation was conducted on the aerodynamic imbalance of the wind turbine rotor, and the second investigation results were obtained. Based on the first and second investigation results, a system analysis of wind turbine vibration and abnormal noise issues was obtained.

2. The system analysis method for wind turbine vibration and abnormal noise problems according to claim 1, characterized in that, Determine the basic fault information of the wind turbine, including: Determine the occurrence mode, location, and conditions of vibration and abnormal noise problems in wind turbine units; The occurrence mode, location, and conditions are used as the basic fault information for wind turbine units.

3. The system analysis method for wind turbine vibration and abnormal noise problems according to claim 1, characterized in that, The multidimensional measured data includes: raw vibration data during the period when the wind turbine failure occurred; Online vibration data analysis of the wind turbine generator includes: The original vibration data were subjected to time-domain feature analysis and frequency-domain feature analysis respectively to obtain online vibration characteristic parameters; The online vibration characteristic parameters are compared with the reference vibration characteristic parameters obtained in advance during historical normal operation periods to obtain the first comparison result; Based on the first comparison result and combined with the basic fault information, the damaged component and the degree of damage to the damaged component are determined, and the online vibration data analysis results are obtained.

4. The system analysis method for wind turbine vibration and abnormal noise problems according to claim 1, characterized in that, The multidimensional measured data includes: second-level operating data within a set time period before and after the wind turbine failure; The wind turbine generator set is subjected to unit operation data analysis, including: The second-level operating data of the wind turbine unit within a set time period before and after the fault is compared with the normal operating data of other units in the same period obtained in advance to obtain a second comparison result; Based on the second comparison result, the operational changes of the wind turbine unit before and after the fault and other normal units in the same time period are analyzed. Combined with the basic fault information, the abnormal points in the wind turbine unit are preliminarily determined, and the unit operation data analysis results are obtained.

5. The system analysis method for wind turbine vibration and abnormal noise problems according to claim 4, characterized in that, The second-level operating data includes: real-time wind speed, operating speed, active power, main shaft gearbox temperature, generator bearing temperature, oil temperature, pitch angle, pitch speed, and yaw system pressure.

6. The system analysis method for vibration and abnormal noise problems of wind turbine units according to claim 1, characterized in that, The multidimensional measured data includes: historical inspection and maintenance records of problematic components in the wind turbine; The maintenance and repair records of the wind turbine units were analyzed, including: Based on the historical maintenance records, the wind turbine units were located and the locations of frequently faulty components were identified to determine the areas where historical faults were concentrated. Based on the aforementioned basic fault information, the analysis of influencing factors, diagnosis of the rationality of solutions, quality inspection and acceptance verification, and verification of key assembly dimensions are conducted on the areas where historical faults are concentrated, resulting in the analysis results of maintenance records.

7. The system analysis method for wind turbine vibration and abnormal noise problems according to claim 1, characterized in that, Based on the preliminary analysis results, a layered inspection of the mechanical components of the wind turbine was conducted, resulting in the first inspection results, including: Based on the preliminary analysis results, the mechanical components of the wind turbine were inspected at the foundation layer, tower layer, and nacelle interior and related components, respectively, and the inspection sub-results for each layer were obtained. The investigation results corresponding to each layer are summarized to obtain the first investigation result.

8. The system analysis method for wind turbine vibration and abnormal noise problems according to claim 7, characterized in that, The mechanical components of the wind turbine were inspected, including the interior of the nacelle and related parts. The troubleshooting should proceed in order from the generator to the impeller, starting with the closest to the furthest point. Specifically, this includes: The generator, coupling, gearbox, main shaft, nacelle, yaw system, pitch system, and impeller were inspected in sequence.

9. The system analysis method for vibration and abnormal noise problems of wind turbine units according to claim 1, characterized in that, Based on the preliminary analysis results, a special investigation was conducted on the aerodynamic imbalance of the wind turbine rotor, resulting in the following second investigation results: Obtain measured images of the impeller; Based on the measured images of the impeller, the actual distance from the blade tip to the tower is determined, a trend diagram of the net clearance change of the three blades is generated, and the degree of imbalance is determined according to the vibration velocity classification. Based on the aforementioned imbalance and the preliminary analysis results, the aerodynamic imbalance influencing factors of the impeller were investigated, and the second investigation results were obtained.

10. The system analysis method for vibration and abnormal noise problems of wind turbine units according to claim 9, characterized in that, Based on the aforementioned imbalance and the preliminary analysis results, the aerodynamic imbalance influencing factors of the impeller were investigated, resulting in a second investigation result, including: Based on the aforementioned imbalance and the preliminary analysis results, the blade mass moment deviation was checked, the blade electrical zero position was calibrated, the deviation between the zero-degree scale and the positioning block was detected, the yaw positioning accuracy was investigated, and the pitch system coordination was tested to obtain the second investigation results.

Images