A wind turbine variable pitch bearing damage simulation device and method

Abstract

The application discloses a wind turbine variable pitch bearing damage simulation device and method, the device comprises a wind turbine simulation test bench, a wear type damage simulation unit and a signal acquisition unit, the wear type damage simulation unit comprises a bearing sleeve, a shaft, two angular contact ball bearings, an axial adjusting piece and a circumferential locking assembly; by replacing the axial adjusting piece arranged between the inner rings of the two angular contact ball bearings, the increase of the clearance caused by the distributed wear of the variable pitch bearing, the attenuation of the pre-tightening and the change of the supporting state are simulated equivalently, the controllability and repeatability of the test state are improved; the circumferential locking assembly limits the relative rotation of the shaft and the bearing sleeve in the variable pitch direction, the commutation transient vibration response under different supporting states is obtained, and the variable pitch movement is not limited; without using large real variable pitch bearings, irreversible damage processing is not needed for the test bearings, and the test cost is reduced.

Description

Technical Field

[0001] This invention belongs to the field of wind turbine bearing damage simulation and vibration testing technology, specifically relating to a wind turbine pitch bearing damage simulation device and method. Background Technology

[0002] With the large-scale development of offshore wind power, the single-unit capacity, blade length, and hub height of wind turbines are constantly increasing, and the loads borne by the blades and their connecting support components are also increasing accordingly. The pitch bearing, installed between the blade root and the hub, bears radial loads, axial loads, and overturning moments, and must also meet the blade pitch angle adjustment requirements, making it a key support component in the pitch system. During operation, the pitch bearing continuously revolves around the wind turbine's main shaft along with the blades and hub, but it only performs low-speed, limited-angle reciprocating pitch changes during power regulation, shutdown, or safety control. When wind speed and power demand are relatively stable, the pitch bearing typically remains stationary for extended periods. Its heavy-load, low-speed, intermittent start-stop, and static load characteristics differ significantly from those of conventional continuously rotating rolling bearings.

[0003] Existing rolling bearing damage simulation and fault diagnosis methods are mostly based on the excitation of local defects. This involves machining localized damage such as pitting, dents, or spalling on the raceway or rolling element surface, and then continuously rotating the inner and outer rings of the bearing relative to each other, causing the defects to repeatedly enter the load-bearing area and generate speed-related periodic impacts. For pitch bearings, the pitch motion is characterized by low speed, limited stroke, and frequent starts and stops. Localized damage cannot continuously pass through the load-bearing area according to a fixed cycle. Furthermore, the rolling elements may experience sliding and unstable contact during start-up, stopping, and reversing. Therefore, the applicability of traditional fault characteristic frequency methods for the discontinuous motion conditions of pitch bearings is significantly limited.

[0004] After long-term service, pitch bearings may experience localized damage such as pitting and spalling, as well as significant wear in the contact area between the rolling elements and raceways. This type of wear primarily alters the internal contact geometry and overall support state of the bearing, manifesting as increased internal clearance, weakened preload, decreased support stiffness, and changes in the load-bearing area distribution. This differs from the periodic impacts caused by repeated passage of localized defects through the load-bearing area. As the pitch bearing revolves around the main shaft with the blades and hub, the blade orientation relative to the direction of gravity continuously changes. The axial load, radial load, and overturning moment exerted by the blade weight on the pitch bearing also change periodically with the rotor orientation. Within one rotor revolution cycle, the corresponding load components undergo directional reversals, shifting the primary load-bearing position of the pitch bearing from one side to the other, and causing load path reversals and contact load redistribution. The superposition of aerodynamic and inertial loads alters the load amplitude and reversal timing, further complicating the actual load state of the pitch bearing.

[0005] When wear leads to increased clearance, weakened preload, or decreased support stiffness, load reversal causes the rolling elements on the original bearing side to gradually unload, while the rolling elements on the other side re-enter the bearing position, resulting in a transfer of contact load between different bearing paths. The clearance closure and re-loading process can generate strong, non-stationary commutation transient impacts. This response is jointly formed by the change in the direction of the rotor's revolution load and the post-wear support state, independent of the continuous relative rotation of the inner and outer rings of the pitch bearing. This provides a different experimental approach than traditional methods of excitation for studying wear-type damage in pitch bearings under static holding or finite angular oscillation conditions.

[0006] However, conducting wear and damage tests directly using large, real pitch bearings requires large-sized specimens, specialized mounting structures, and multi-directional loading equipment, and involves transportation, hoisting, and safety protection, resulting in high testing costs. When using retired bearings, their service history, wear degree, and damage distribution are difficult to control, and there is a lack of consistency between different specimens; damage processed on real bearings is irreversible, which is not conducive to repeated switching and comparison of different states. Although existing small bearing testing devices can reduce costs, they mostly use continuous rotation or local defect excitation methods, or only simulate clearance changes by changing the assembly spacing of ordinary bearings, which cannot simultaneously reflect the bidirectional load relationship of pitch bearings, the load transmission path of blade-bearing-hub, the constraint conditions of no normal pitch relative motion, and the reversing condition of wind turbine revolution load. Therefore, there is an urgent need for a pitch bearing damage simulation device and method that can construct different wear equivalent states at a lower cost and obtain the transient impact vibration response during blade revolution. Summary of the Invention

[0007] This invention aims to provide a device and method for simulating the damage of pitch bearings in wind turbines, solving the problems of high cost of large-scale real pitch bearing wear tests, difficulty in reproducing wear states, and incompatibility between traditional continuous-pass simulation methods for local defects and the operating characteristics of pitch bearings under low speed, limited angle, and long-term static loading. It achieves low-cost and repeatable simulation of the distributed wear state of pitch bearings and their commutation transient vibration response.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A wind turbine pitch bearing damage simulation device includes a wind turbine simulation test bench, a wear-type damage simulation unit, and a signal acquisition unit. The wind turbine simulation test bench includes a main shaft, a hub, blades, and a drive unit for driving the hub and blades to revolve around the main shaft as a whole. The wear-type damage simulation unit is installed between the hub and the blades. The wear-type damage simulation unit includes a bearing sleeve, a shaft, a first angular contact ball bearing, a second angular contact ball bearing, an outer ring axial positioning assembly, an inner ring axial clamping assembly, an axial adjustment component, and a circumferential locking assembly. The first angular contact ball bearing and the second angular contact ball bearing are arranged along the axial direction of the shaft. The outer ring wide end face of the first angular contact ball bearing and the outer ring wide end face of the second angular contact ball bearing are opposite to each other, that is, the first angular contact ball bearing and the second angular contact ball bearing are arranged back to back along the axial direction of the shaft. The outer rings of the two angular contact ball bearings are positioned in the bearing sleeve by the outer ring axial positioning assembly. The inner rings of the two angular contact ball bearings are installed on the shaft and clamped by the inner ring axial clamping assembly. The axial adjustment element is disposed between the opposite end faces of the inner rings of the two angular contact ball bearings. By replacing the axial adjustment element with different axial thicknesses, the axial positional relationship between the inner ring assembly and the outer ring assembly of the two angular contact ball bearings is changed to simulate the reference support state and at least one wear equivalent support state. Different support states correspond to different axial free displacements, preload states, or equivalent support stiffness, which are used to equivalently characterize the increase in clearance, preload attenuation, and changes in support state caused by distributed wear of the pitch bearing. The bearing sleeve is connected to the hub via a second bearing end cap. The second bearing end cap has an externally threaded connecting shaft extending outward along the axial direction. The externally threaded connecting shaft is threadedly connected to the hub. The end of the shaft facing the blade has an internally threaded hole. The blade has an externally threaded connector that mates with the internally threaded hole. The externally threaded connector is screwed into the internally threaded hole to make the shaft and the blade detachably connected. The circumferential locking assembly is used to restrict the circumferential relative rotation of the shaft with respect to the bearing sleeve around the axis of the shaft, so that the wear-type damage simulation unit revolves with the hub and the blade as a whole without pitch relative rotation. When the drive unit drives the hub and the blade to revolve, at least one of the following load changes is satisfied: at least one load component of the blade's self-weight in the local coordinate system rotating with the wear-type damage simulation unit changes periodically with the wind turbine's orientation, and changes from a positive value to a negative value or from a negative value to a positive value within one revolution period; or the direction of the combined load formed by the blade's self-weight and at least one of aerodynamic and inertial loads changes with the wind turbine's orientation; the load change causes a redistribution of the contact load and bearing path of the paired bearings, and generates different commutation transient vibration responses under different support conditions; The signal acquisition unit is used to acquire the vibration response of at least one of the bearing sleeve, the hub, and the main shaft during the overall revolution of the hub and the blade.

[0009] The above-mentioned solution provided by the present invention can construct different wear equivalent support states and obtain the commutation transient vibration response under the condition of changing load direction without processing local damage on the raceway or rolling elements of a large real pitch bearing, or by causing the blade connection end and the hub connection end to continuously rotate relative to each other.

[0010] According to embodiments of the present invention, the present invention can be further optimized, and the following are the optimized technical solutions.

[0011] In one preferred embodiment, the axial adjustment element is an annular adjustment shim.

[0012] In one preferred embodiment, the outer ring axial positioning assembly includes an axial positioning surface formed within the bearing sleeve and a first bearing end cap mounted on one end of the bearing sleeve, wherein the outer rings of the two angular contact ball bearings are axially defined between the axial positioning surface and the first bearing end cap; the inner ring axial clamping assembly includes a lock nut and a retaining washer mounted on the shaft, wherein the lock nut and the retaining washer are used to axially clamp the inner ring assembly of the two angular contact ball bearings.

[0013] In one preferred embodiment, the circumferential locking assembly includes at least one circumferential limiting member disposed between the shaft and the bearing sleeve, the circumferential limiting member restricting the circumferential relative rotation of the shaft with respect to the bearing sleeve about the axis of the shaft by form fit. In one preferred embodiment, the circumferential limiting member includes an adjusting bracket mounted on the bearing sleeve and an I-beam locating pin that cooperates with the adjusting bracket; the I-beam locating pin cooperates with the shaft to restrict the shaft from rotating circumferentially relative to the bearing sleeve about the axis of the shaft.

[0014] In one preferred embodiment, the shaft, the first angular contact ball bearing, the second angular contact ball bearing, and the replaceable axial adjustment element are combined into a detachable module that can be installed or removed from the bearing sleeve as a whole.

[0015] In one preferred embodiment, the signal acquisition unit is a phase acquisition component for acquiring the real-time wind turbine azimuth of the hub or the blade relative to the main shaft; the signal acquisition unit is connected to a data processing unit; the data processing unit is used to determine the characteristic range in which the sign of the at least one load component changes or the direction of the resultant load changes, based on the correspondence between the real-time wind turbine azimuth, the preset wind turbine azimuth and the sign of at least one load component or the direction of the resultant load.

[0016] Based on the same inventive concept, this invention also provides a method for simulating damage to wind turbine pitch bearings, including: S1, Under the same inner ring axial clamping process, the axial adjustment parts with different axial thicknesses are selected respectively to change the axial position relationship between the inner ring group and the outer ring group of the first angular contact ball bearing and the second angular contact ball bearing, so as to simulate the reference support state and at least one wear equivalent support state. S2, The wear-type damage simulation unit is installed between the hub and the blade, and the circumferential locking component restricts the circumferential relative rotation of the shaft relative to the bearing sleeve around the axis of the shaft, so that the wear-type damage simulation unit revolves with the hub and the blade as a whole without pitch relative rotation. S3, in both the reference support state and the wear equivalent support state, the drive unit drives the hub and blades to revolve around the main shaft as a whole, ensuring that at least one of the following load changes is satisfied during the revolution: The first type of load change: at least one load component of the blade's self-weight in the local coordinate system rotating with the wear-type damage simulation unit changes periodically with the wind turbine's orientation, and changes from a positive value to a negative value or from a negative value to a positive value within one revolution period; The second type of load change: the direction of the combined load formed by the blade's own weight and at least one of aerodynamic and inertial loads changes with the wind turbine's orientation; S4. Under the same impeller speed and main shaft load settings, the vibration response of at least one of the bearing sleeve, hub and main shaft under the reference support state and the wear equivalent support state is collected respectively.

[0017] In one preferred embodiment, it further includes: S5. Based on the correspondence between the real-time wind turbine orientation, the preset wind turbine orientation, and the sign of the at least one load component or the direction of the resultant load, determine the characteristic range in which the sign of the at least one load component changes or the direction of the resultant load changes. The correspondence between the preset wind turbine orientation and the sign of the at least one load component is determined based on the geometric relationship between the blade mass, the blade centroid position, the blade installation posture, the local coordinate system, and the wind turbine orientation. The correspondence between the preset wind turbine orientation and the direction of the combined load is determined based on the preset load model.

[0018] In one preferred embodiment, it further includes: S6. Extract the vibration response time window containing the feature interval and the preset time length before and after the feature interval, extract the commutation transient impact features in the vibration response time window, and compare the difference in commutation transient impact features between the reference support state and the wear equivalent support state under the same wind turbine speed and main shaft load settings. The differences in the transient impact characteristics of the reversal are characterized by at least one of the vibration peak value, peak-to-peak value, kurtosis, envelope peak value, and preset frequency band energy.

[0019] In one preferred embodiment, it further includes: S7. After completing the vibration response acquisition under a wear equivalent support state, stop the revolution of the hub and the blade, remove the first bearing end cover and the second bearing end cover, release the circumferential lock between the shaft and the bearing sleeve and the axial clamping of the inner ring assemblies of the two angular contact ball bearings, remove the detachable module consisting of the shaft, the first angular contact ball bearing, the second angular contact ball bearing and the axial adjustment component from the bearing sleeve, replace the axial adjustment component, re-clamp the inner ring assemblies of the two angular contact ball bearings using the same inner ring axial clamping process, reinstall the detachable module into the bearing sleeve and restore the circumferential lock, so as to realize the switching between the reference support state and the wear equivalent support state or between different wear equivalent support states.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention provides a device and method for simulating damage to pitch bearings of wind turbine units. By replacing the axial adjustment component set between the inner rings of two angular contact ball bearings, the axial positional relationship between the inner ring group and the outer ring group is changed, forming a reference support state and different wear equivalent support states. This effectively simulates the increase in clearance, preload attenuation, and changes in support state caused by distributed wear of the pitch bearing, improving the controllability and repeatability of the test state. The present invention restricts the relative rotation of the shaft and bearing sleeve in the pitch direction through a circumferential locking component, while allowing the wear-type damage simulation unit to revolve with the hub and blades as a whole. By utilizing the change in load component or the change in the direction of the resultant load of the blade load with the wind turbine azimuth, the commutation transient vibration response under different support states is obtained. It does not rely on the transient impact response caused by damage contact during the relative movement between the outer ring, inner ring, and rolling elements of the bearing, and is therefore not limited by pitch motion. This invention eliminates the need for large, real pitch bearings and irreversible damage processing of the test bearings. By replacing the axial adjustment components, it enables repeated switching between different wear-equivalent support states, reducing testing costs and providing test conditions and data support for the response mechanism analysis and detection method verification of wear-type damage in wind turbine pitch bearings. Attached Figure Description

[0021] Figure 1 It shows the axial force, radial force, and overturning moment caused by the blades; Figure 2 This is a diagram showing the load distribution on the blades; Figure 3 This is a cross-sectional view of a pitch bearing damage simulation device according to an embodiment of the present invention; Figure 4This is a perspective view of a pitch bearing damage simulation device according to an embodiment of the present invention; Figure 5 This is a front view of a pitch bearing damage simulation device according to an embodiment of the present invention; Figure 6 This is a side view of a pitch bearing damage simulation device according to an embodiment of the present invention; Figure 7 This is a diagram of a wind turbine simulation test bench and test site according to an embodiment of the present invention; Figure 8 This is a solid drawing of the adjusting shim, angular contact bearing, lock nut, and shaft in a pitch bearing damage simulation device according to an embodiment of the present invention. Figure 9 This is a vibration acceleration signal of a pitch bearing under simulated healthy and damaged conditions according to an embodiment of the present invention. Figure 9 (a) is the radial signal. Figure 9 (b) is the tangential signal. Figure 9 (c) represents the axial signal; Figure 10 This is a partially enlarged view of the vibration acceleration signal of a pitch bearing under simulated healthy and damaged conditions according to an embodiment of the present invention. Figure 10 (a) is the radial signal. Figure 10 (b) is the tangential signal. Figure 10 (c) represents the axial signal; Figure 11 This is a diagram illustrating the separation and spectral analysis process of tangential vibration signals from a pitch bearing under simulated healthy and damaged conditions, according to an embodiment of the present invention. Figure 11 (a) is a graph of the original test signals for a healthy state. Figure 11 (b) is a diagram of the original test signal under damage conditions. Figure 11 (c) is a diagram of the filtered signal for the healthy state. Figure 11 (d) is a diagram of the filtered signal under damage conditions. Figure 11 (e) is the envelope signal diagram of the health status. Figure 11 (f) is the envelope signal diagram of the damage state. Figure 11 (g) is the squared envelope spectrum of the healthy state. Figure 11 (f) is the squared envelope spectrum of the damage state; Among them, 1 is a fastener, 2 is the first bearing end cover, 3 is a lock nut, 4 is a stop washer, 5 is a bearing sleeve, 61 is the first angular contact bearing, 62 is the second angular contact bearing, 7 is an adjusting bracket, 8 is an I-beam locating pin, 9 is the second bearing end cover, 10 is a shaft, 11 is an annular adjusting shim, 21 is a planetary gearbox, 22 is the main shaft, 23 is a blade, 24 is a wear-type damage simulation device, 25 is a vibration sensor, 26 is a hub, 27 is the main shaft bearing housing, 28 is an electromagnetic load, 29 is a blower, 30 is a pitch bearing, 31 is a pinion, and 32 is a pitch motor. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0023] Example 1 This invention provides a device and method for simulating damage to wind turbine pitch bearings. Under conditions where large real pitch bearings cannot be used, raceways are not machined, or rolling elements are partially damaged, a reference support state and a wear-equivalent support state are constructed, and the vibration response under different support states during the change of blade load direction is obtained.

[0024] like Figures 3 to 6 As shown, the damage simulation unit includes fastener 1, first bearing end cover 2, lock nut 3, stop washer 4, bearing sleeve 5, first angular contact ball bearing 61, second angular contact ball bearing 62, adjusting bracket 7, I-beam positioning pin 8, second bearing end cover 9, shaft 10, and annular adjusting shim 11 as a replaceable axial adjusting component.

[0025] The first angular contact ball bearing 61 and the second angular contact ball bearing 62 are arranged back-to-back in an O-shape along the axial direction of shaft 10, meaning that the pressure lines of the two angular contact ball bearings flare outwards to both sides of the bearing assembly along the axial direction. The two single-row angular contact ball bearings serve as paired support components for the wear-type damage simulation unit, forming a support structure capable of transmitting radial loads, axial loads, and overturning moments.

[0026] It should be noted that the use of back-to-back single-row angular contact ball bearings is the basic support method for constructing the wear-type damage simulation unit in this embodiment of the invention. This embodiment of the invention does not replicate the raceway structure and contact geometry of a large double-row four-point contact ball pitch bearing on a proportional scale, but rather utilizes paired bearings to construct an adjustable equivalent support state to study the changes in support state caused by distributed wear and its vibration response.

[0027] The bearing sleeve 5 has an axial positioning surface. The outer rings of the first angular contact ball bearing 61 and the second angular contact ball bearing 62 are disposed within the bearing sleeve 5 and are axially constrained between the axial positioning surface and the first bearing end cap 2. The first bearing end cap 2 is detachably connected to the bearing sleeve 5 by fasteners 1, thereby achieving axial positioning of the outer rings of the two single-row angular contact ball bearings.

[0028] The inner rings of the first angular contact ball bearing 61 and the second angular contact ball bearing 62 are mounted on the shaft 10. A lock nut 3 and a locking washer 4 are provided on the shaft 10 to axially clamp the inner rings of the two single-row angular contact ball bearings according to a predetermined locking process, and to prevent the lock nut 3 from loosening during testing.

[0029] An annular adjusting shim 11 is disposed between the opposing end faces of the inner rings of the first angular contact ball bearing 61 and the second angular contact ball bearing 62. By selecting annular adjusting shims 11 with different axial thicknesses, the axial positional relationship between the inner ring groups of the two single-row angular contact ball bearings and their outer ring groups is changed, thereby forming a reference support state and at least one wear-equivalent support state.

[0030] The annular adjusting shims 11 of different thicknesses are adjustment parameters for the support state. Their thickness is not directly equivalent to the actual axial free displacement after the mating bearings are assembled, nor is it directly equivalent to the internal clearance of the actual pitch bearing. The state formed after assembly can be characterized by at least one of the following: the axial free displacement of the shaft 10 relative to the bearing sleeve 5, the load-displacement stiffness within a predetermined axial load range, and the preload force.

[0031] Specifically, under the same conditions of outer ring positioning method, inner ring locking process and other assembly conditions, the support state with small axial free displacement, high preload or large load-displacement stiffness is set as the reference support state; the support state with increased axial free displacement, decreased preload or reduced load-displacement stiffness is set as the wear equivalent support state, so as to equivalently characterize the increase in internal clearance, preload attenuation and support state changes caused by distributed wear of pitch bearing.

[0032] The adjusting bracket 7 is mounted on the bearing sleeve 5. The I-beam locating pin 8 cooperates with the adjusting bracket 7 and the shaft 10 to form a circumferential locking assembly between the shaft 10 and the bearing sleeve 5. The circumferential locking assembly restricts the circumferential relative rotation of the shaft 10 with respect to the bearing sleeve 5 around the axis of the shaft 10 through form fit, thereby restricting the relative rotation of the wear-type damage simulation unit in the normal pitch direction.

[0033] The second bearing end cap 9 is detachably mounted on the other end of the bearing sleeve 5. The second bearing end cap 9 is provided with an externally threaded connecting shaft extending outward along the axial direction for connection with the hub 26 of the wind turbine simulation test bench. The end of the shaft 10 facing the blade 23 is provided with an internally threaded hole, and the connecting piece of the blade 23 is provided with an externally threaded connecting section that mates with the internally threaded hole, thereby allowing the shaft 10 and the blade 23 to be detachably connected.

[0034] Through the above connection method, shaft 10 is connected to blade 23, and bearing sleeve 5 is connected to hub 26. Under the locking action of adjusting bracket 7 and I-beam positioning pin 8, there is no circumferential relative rotation between shaft 10 and bearing sleeve 5 in the pitch direction, but the wear-type damage simulation unit can revolve around main shaft 22 as a whole with hub 26 and blade 23.

[0035] like Figure 7 As shown, the wind turbine simulation test bench includes a planetary gearbox 21, a main shaft 22, blades 23, a wear-type damage simulation unit 24, a vibration sensor 25, a hub 26, a main shaft bearing housing 27, an electromagnetic load 28, and a blower 29.

[0036] Blower 29 provides airflow to drive blades 23 and hub 26 to revolve around main shaft 22; electromagnetic load 28 applies load to main shaft 22. Wear-type damage simulation unit 24 is installed between blades 23 and hub 26, and the load generated by blades 23 is transmitted to hub 26 via shaft 10, first angular contact ball bearing 61 and second angular contact ball bearing 62, and bearing sleeve 5.

[0037] As the hub 26 and blades 23 revolve around the main shaft 22, the wear-type damage simulation unit 24 rotates along with the entire unit. As the wind turbine's orientation changes, the radial, tangential, or axial load components of the blade's self-weight in the local coordinate system rotating with the wear-type damage simulation unit 24 change accordingly, with at least one load component changing from a positive value to a negative value or from a negative value to a positive value within one revolution cycle.

[0038] When the blade's own weight acts together with aerodynamic forces and inertial loads, the direction of the resultant load can change with the wind turbine's orientation. When the sign of the load components or the direction of the resultant load changes, the contact loads acting on the first angular contact ball bearing 61 and the second angular contact ball bearing 62 are redistributed accordingly.

[0039] Under the wear equivalent support state, the axial free displacement, preload, or equivalent support stiffness of the paired bearings are different from those under the reference support state. Therefore, during the unloading of the original load path and the reloading of another load path, a commutation transient vibration response different from that under the reference support state can be generated.

[0040] The specific procedures for damage simulation testing are as follows: Step 1: Construct the reference support state and the wear equivalent support state. While keeping the first angular contact ball bearing 61, the second angular contact ball bearing 62, the shaft 10, the bearing sleeve 5, the outer ring positioning method, and the inner ring axial clamping process the same, annular adjusting shims 11 with different axial thicknesses are selected respectively, and the annular adjusting shims 11 are placed between the opposite end faces of the inner rings of the two single-row angular contact ball bearings to change the axial positional relationship between the inner ring group and the outer ring group of the two single-row angular contact ball bearings, forming a reference support state and at least one wear equivalent support state.

[0041] The annular adjusting shims 11 of different thicknesses serve as adjustment parameters for the support state. Their thickness is not directly equivalent to the actual axial free displacement after the mating bearings are assembled or the internal clearance of the actual pitch bearing. The resulting different support states can be characterized by at least one of the following: the axial free displacement of the shaft 10 relative to the bearing sleeve 5, the load-displacement stiffness within the preset axial load range, and the preload.

[0042] The installation process of the damage simulation unit is as follows: Install the second angular contact ball bearing 62 onto the shaft 10; fit the selected annular adjusting shim 11 onto the shaft 10, ensuring that the annular adjusting shim 11 is adjacent to the inner ring end face of the second angular contact ball bearing 62; install the first angular contact ball bearing 61 and the second angular contact ball bearing 62 onto the shaft 10 in an O-ring back-to-back configuration, with the annular adjusting shim 11 positioned between the opposing end faces of the inner rings of the two single-row angular contact ball bearings; install the assembly formed by the shaft 10, the first angular contact ball bearing 61, the second angular contact ball bearing 62, and the annular adjusting shim 11 into the bearing sleeve 5; and install the retaining washer 4 and the locking nut. 3. Install the inner ring assembly of the two single-row angular contact ball bearings on the shaft 10 and clamp them axially according to the preset locking process; install the adjusting bracket 7 on the bearing sleeve 5 and cooperate the I-beam locating pin 8 with the adjusting bracket 7 and the shaft 10 to restrict the circumferential relative rotation of the shaft 10 with respect to the bearing sleeve 5 around the axis; install the second bearing end cover 9 to close and limit the end where the adjusting bracket 7 and the I-beam locating pin 8 are located; install the first bearing end cover 2 and connect it to the bearing sleeve 5 through the fastener 1, so that the first bearing end cover 2 and the axial locating surface inside the bearing sleeve 5 together limit the axial position of the outer ring of the two single-row angular contact ball bearings.

[0043] Step 2: Install the damage simulation unit on the wind turbine simulation test bench. The assembled damage simulation unit is installed between the hub 26 and the blade 23, so that the second bearing end cover 9 is connected to the hub 26 and the shaft 10 is connected to the blade 23.

[0044] After installation, check and adjust the fit between the card holder 7 and the I-beam positioning pin 8 to ensure that the shaft 10 cannot rotate circumferentially relative to the bearing sleeve 5 around the axis of the shaft 10, so that the damage simulation unit revolves as a whole with the hub 26 and the blade 23 without pitch relative rotation.

[0045] Step 3: Conduct vibration response testing The blower 29 is turned on to drive the blades 23, hub 26, and main shaft 22 to rotate, and the main shaft load is set through the electromagnetic load 28. Under the same or comparable impeller speed, main shaft load, sensor installation position, sampling frequency, and sampling duration, the vibration response is measured under the reference support state and the wear equivalent support state, respectively.

[0046] In this embodiment of the invention, a triaxial vibration accelerometer 25 is used to collect vibration acceleration signals of the wheel hub 26 in the radial, tangential, and axial directions. The signal sampling frequency is 8000 Hz, and the duration of a single measurement is 100 s. The obtained vibration signals are as follows: Figure 9 As shown.

[0047] Figure 9 In the text, 0.5Cls indicates the wear-equivalent support state formed by using an annular adjusting shim with a nominal axial thickness of 0.5 mm in the test. This mark is used to distinguish different adjusting component configurations and does not indicate that the actual internal clearance of the paired bearings after assembly is 0.5 mm.

[0048] Depend on Figure 9 It can be seen that, compared with the reference support state, the vibration signal under the wear equivalent support state shows more obvious periodic transient impacts, among which the transient impacts in the tangential vibration signal are more prominent.

[0049] The vibration signal within the range of 18–20 s was locally amplified, and the results are as follows: Figure 10 As shown. By Figure 10 It can be seen that the vibration signal under the wear equivalent support state exhibits relatively obvious transient impacts around 18.12 s, 18.88 s and 19.60 s, while the transient response at the corresponding times under the reference support state is weaker.

[0050] The aforementioned transient impacts are related to the periodic changes in the load direction during the revolution of hub 26 and blade 23. Under the wear-equivalent support state, the axial free displacement, preload, or equivalent support stiffness of the paired bearings change relative to the reference support state, causing a more pronounced transient vibration response during the redistribution of contact load and the conversion of load paths between the two single-row angular contact ball bearings.

[0051] The transient vibration response can be used to characterize the unloading, clearance closure, and reloading processes of paired bearings during changes in the load direction. When the axial free displacement is large, these processes may involve local contact separation and re-contact, but the specific contact state can be further determined by combining the wind turbine phase, support condition measurement results, or contact force analysis.

[0052] In embodiments equipped with a phase acquisition component, the real-time wind turbine azimuth of the hub 26 or blade 23 relative to the main shaft 22 can be acquired synchronously. Based on the correspondence between the preset wind turbine azimuth and the sign of the target load component or the direction of the resultant load, the characteristic range in which the sign of the load component changes or the direction of the resultant load changes is determined. The correspondence between the preset wind turbine azimuth and the sign of the at least one load component is determined based on the geometric relationship between blade mass, blade centroid position, blade installation attitude, local coordinate system, and wind turbine azimuth; the correspondence between the preset wind turbine azimuth and the direction of the resultant load is determined based on a preset load model.

[0053] Step 4: Perform signal separation and feature extraction This invention takes tangential vibration signal as an example to perform bandpass filtering and envelope demodulation on the vibration response under the reference support state and the wear equivalent support state.

[0054] The original test signals under the reference support condition and the wear equivalent support condition are as follows: Figure 11 (a) and Figure 11 As shown in (b). The bandpass filter frequency band was set to 280–450 Hz, and the original test signal was bandpass filtered to obtain... Figure 11 (c) and Figure 11 The filtered signal shown in (d) is subjected to envelope demodulation to obtain... Figure 11 (e) and Figure 11 (f) shows the envelope signal.

[0055] Depend on Figure 11 It can be seen that there are obvious periodic transient impact components in the filtered signal and envelope signal under the wear equivalent support state, and the maximum value of the transient impact is close to 0.5g, where g is the gravitational acceleration, which is about 9.8 m / s². No periodic transient impact of the same degree is observed under the reference support state, and the amplitude of the filtered signal and envelope signal is generally lower than 0.05g.

[0056] right Figure 11 (e) and Figure 11 The squared envelope spectrum analysis of the envelope signal shown in (f) yielded the following results: Figure 11 (g) and Figure 11 As shown in (h).

[0057] Depend on Figure 11(g) and Figure 11 (h) It can be seen that under the wear-equivalent support state, the 2nd, 4th, 6th, and 8th harmonic components with the transient impact repetition frequency as the fundamental frequency are more prominent, and the amplitudes shown in the figure are respectively g、 g、 g and g, even the 10th and 12th harmonics are clearly visible. In contrast, these obvious characteristic frequency components are not present in the healthy state, and the maximum value of the squared envelope spectrum in the healthy state is [value missing]. g is much smaller than 1 / 10 of the characteristic frequency components under fault conditions.

[0058] The unit of the square envelope spectrum amplitude should be in [unit]. Figure 11 The definition of the vertical axis and the corresponding data processing method shall prevail. When the square envelope spectrum is not normalized, its unit should not be directly expressed as the vibration acceleration unit g.

[0059] The above results show that by replacing the annular adjusting shim 11 set between the inner rings of the two single-row angular contact ball bearings, the support state of the paired bearings can be changed; under the condition that the shaft 10 and the bearing sleeve 5 do not rotate relative to each other and the damage simulation unit revolves as a whole with the hub 26 and the blade 23, the reference support state and the wear equivalent support state produce distinguishable transient vibration responses during the change of the load direction.

[0060] The specific process of feature extraction is as follows: extract the vibration response time window containing the feature interval and a preset time length before and after the feature interval; extract the commutation transient impact features in the vibration response time window; and compare the differences in commutation transient impact features between the reference support state and the wear equivalent support state under the same wind turbine speed and main shaft load settings; wherein, the differences in commutation transient impact features are characterized by at least one of vibration peak value, peak-to-peak value, kurtosis, envelope peak value and preset frequency band energy.

[0061] Step 5: Switch support status After completing the vibration response acquisition under a support condition, stop the revolution of the blower 29, hub 26 and blade 23, and remove the damage simulation unit from between hub 26 and blade 23.

[0062] The disassembly process of the damage simulation unit is as follows: Remove the second bearing end cover 9, remove the I-beam locating pin 8 and the adjusting bracket 7, and release the circumferential lock between the shaft 10 and the bearing sleeve 5; remove the first bearing end cover 2, remove the locking nut 3 and the locking washer 4, and release the axial clamping of the inner ring assemblies of the two single-row angular contact ball bearings; remove the assembly formed by the shaft 10, the first angular contact ball bearing 61, the second angular contact ball bearing 62 and the annular adjusting shim 11 from the bearing sleeve 5 as a whole; remove the first angular contact ball bearing 61 and the annular adjusting shim 11 from the shaft 10.

[0063] Shaft 10, first angular contact ball bearing 61, second angular contact ball bearing 62, and annular adjusting shim 11 can be installed or removed from bearing sleeve 5 as a whole as a detachable module. After replacing the annular adjusting shim 11 with another axial thickness specification, it can be reassembled according to the same installation sequence and locking process to achieve switching between the reference support state and the wear equivalent support state or between different wear equivalent support states.

[0064] The above experiments demonstrate that the embodiments of the present invention can construct support states with different axial free displacements, preload degrees, or equivalent support stiffness without machining the bearing raceways or actual damage to the rolling elements, and obtain the corresponding transient vibration response during the change of load direction, thus verifying the feasibility of the wind turbine pitch bearing wear-type damage equivalent simulation device and method provided by the embodiments of the present invention.

[0065] The above embodiments are only used to more clearly illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Equivalent substitutions, simple modifications, or improvements made by those skilled in the art to the above embodiments without departing from the concept of the present invention should all fall within the scope of protection defined by the claims of the present invention.

Claims

1. A device for simulating damage to the pitch bearing of a wind turbine, characterized in that, It includes a wind turbine simulation test bench, a wear-type damage simulation unit, and a signal acquisition unit; The wind turbine simulation test bench includes a main shaft (22), a hub (26), blades (23), and a drive unit for driving the hub (26) and the blades (23) to revolve around the main shaft (22) as a whole. The wear-type damage simulation unit is installed between the hub (26) and the blades (23). The wear-type damage simulation unit includes a bearing sleeve (5), a shaft (10), a first angular contact ball bearing (61), a second angular contact ball bearing (62), an outer ring axial positioning assembly, a second bearing end cap (9), an inner ring axial clamping assembly, an axial adjustment component, and a circumferential locking assembly. The first angular contact ball bearing (61) and the second angular contact ball bearing (62) are arranged along the axial direction of the shaft (10). The outer ring wide end face of the first angular contact ball bearing (61) is opposite to the outer ring wide end face of the second angular contact ball bearing (62). The outer rings of the two angular contact ball bearings are positioned in the bearing sleeve (5) by the outer ring axial positioning assembly. The inner rings of the two angular contact ball bearings are installed on the shaft (10) and clamped by the inner ring axial clamping assembly. The axial adjustment element is disposed between the opposite end faces of the inner rings of the two angular contact ball bearings; The bearing sleeve (5) is connected to the hub (26) via a second bearing end cap (9). The second bearing end cap (9) is provided with an externally threaded connecting shaft extending outward along the axial direction. The externally threaded connecting shaft is threadedly connected to the hub (26). The shaft (10) has an internally threaded hole at one end facing the blade (23). The blade (23) is provided with an externally threaded connector that mates with the internally threaded hole. The externally threaded connector is screwed into the internally threaded hole so that the shaft (10) and the blade (23) are detachably connected. The circumferential locking assembly is used to restrict the circumferential relative rotation of the shaft (10) with respect to the bearing sleeve (5) around the axis of the shaft (10). The signal acquisition unit is used to acquire the vibration response of at least one of the bearing sleeve (5), the hub (26) and the main shaft (22) during the overall revolution of the hub (26) and the blade (23).

2. The wind turbine pitch bearing damage simulation device according to claim 1, characterized in that, The axial adjustment component is an annular adjustment shim (11).

3. The wind turbine pitch bearing damage simulation device according to claim 1, characterized in that, The outer ring axial positioning assembly includes an axial positioning surface formed in the bearing sleeve (5) and a first bearing end cap (2) installed at one end of the bearing sleeve (5). The outer rings of the two angular contact ball bearings are axially defined between the axial positioning surface and the first bearing end cap (2). The inner ring axial clamping assembly includes a locking nut (3) and a retaining washer (4) installed on the shaft (10). The locking nut (3) and the retaining washer (4) are used to axially clamp the inner ring assembly of the two angular contact ball bearings.

4. The wind turbine pitch bearing damage simulation device according to claim 1, characterized in that, The circumferential locking assembly includes at least one circumferential limiting member disposed between the shaft (10) and the bearing sleeve (5), the circumferential limiting member restricting the circumferential relative rotation of the shaft (10) with respect to the bearing sleeve (5) about the axis of the shaft (10) by form fit.

5. The wind turbine pitch bearing damage simulation device according to claim 4, characterized in that, The circumferential limiting component includes an adjusting bracket (7) mounted on the bearing sleeve (5) and an I-beam positioning pin (8) cooperating with the adjusting bracket (7); the I-beam positioning pin (8) cooperates with the shaft (10) to restrict the shaft (10) from rotating circumferentially relative to the bearing sleeve (5) around the axis of the shaft (10).

6. The wind turbine pitch bearing damage simulation device according to claim 1, characterized in that, The shaft (10), the first angular contact ball bearing (61), the second angular contact ball bearing (62) and the axial adjustment component are combined to form a detachable module that can be installed or removed from the bearing sleeve (5) as a whole.

7. The wind turbine pitch bearing damage simulation device according to claim 1, characterized in that, The signal acquisition unit is a phase acquisition component used to acquire the real-time wind turbine orientation of the hub (26) or the blade (23) relative to the main shaft (22); the signal acquisition unit is connected to the data processing unit and is used to determine the characteristic range in which the sign of at least one load component changes or the direction of the resultant load changes.

8. A method for simulating damage to the pitch bearing of a wind turbine, characterized in that, include: S1. Under the same inner ring axial clamping process, the axial adjustment parts with different axial thicknesses are selected respectively to change the axial position relationship between the inner ring group of the first angular contact ball bearing (61) and the second angular contact ball bearing (62) relative to the outer ring group, so as to simulate the reference support state and at least one wear equivalent support state. S2. The wear-type damage simulation unit is installed between the hub (26) and the blade (23). The circumferential locking assembly restricts the shaft (10) from rotating circumferentially relative to the bearing sleeve (5) around the axis of the shaft (10), so that the wear-type damage simulation unit revolves as a whole with the hub (26) and the blade (23) without pitch relative rotation. S3. In the reference support state and the wear equivalent support state respectively, the hub (26) and the blade (23) are driven by the drive unit to revolve around the main shaft (22) as a whole, so that at least one of the following load changes is satisfied during the revolution: The first type of load change: at least one load component of the blade's self-weight in the local coordinate system rotating with the wear-type damage simulation unit changes periodically with the wind turbine's orientation, and changes from a positive value to a negative value or from a negative value to a positive value within one revolution period; The second type of load change: the direction of the combined load formed by the blade's own weight and at least one of aerodynamic and inertial loads changes with the wind turbine's orientation; S4. Under the same wind turbine speed and main shaft load settings, collect the vibration response of at least one of the bearing sleeve (5), hub (26) and main shaft (22) under the reference support state and the wear equivalent support state, respectively.

9. The method for simulating damage to wind turbine pitch bearings according to claim 8, characterized in that, Also includes: S5. Based on the correspondence between the real-time wind turbine orientation, the preset wind turbine orientation, and the sign of the at least one load component or the direction of the combined load, determine the characteristic range in which the sign of the at least one load component changes or the direction of the combined load changes.

10. The method for simulating damage to the pitch bearing of a wind turbine according to claim 9, characterized in that, Also includes: S6. Extract the vibration response time window containing the feature interval and the preset time length before and after the feature interval, extract the commutation transient impact features in the vibration response time window, and compare the difference in commutation transient impact features between the reference support state and the wear equivalent support state under the same wind turbine speed and main shaft load settings. The differences in the transient impact characteristics of the reversing are characterized by at least one of the vibration peak value, peak-to-peak value, kurtosis, envelope peak value, and preset frequency band energy.

11. The wind turbine pitch bearing damage simulation method according to claim 8, characterized in that, Also includes: S7. After completing the vibration response acquisition under a support state, stop the revolution of the hub (26) and the blade (23), remove the first bearing end cover (2) and the second bearing end cover (9), release the circumferential lock between the shaft (10) and the bearing sleeve (5) and the axial clamping of the inner ring groups of the two angular contact ball bearings, take out the detachable module consisting of the shaft (10), the first angular contact ball bearing (61), the second angular contact ball bearing (62) and the axial adjustment component from the bearing sleeve (5), replace the axial adjustment component, re-clamp the inner ring groups of the two angular contact ball bearings using the same inner ring axial clamping process, reinstall the detachable module into the bearing sleeve (5) and restore the circumferential lock.

Images