Bearing degradation detection method, device and equipment of wind driven generator and storage medium
By calculating real-time load and using a temperature prediction model to compare actual and theoretical temperatures, the method addresses environmental interference in wind turbine bearing detection, ensuring high accuracy and cost-effectiveness.
Patent Information
- Application Number
- CN202411018862.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-12-06
AI Technical Summary
Existing generator bearing degradation detection in wind turbines is affected by environmental disturbances, leading to low detection accuracy.
Collect wind field data and bearing load data to calculate real-time load, use a bearing temperature prediction model to predict theoretical temperature, and compare it with actual temperature to determine degradation.
This method provides high accuracy and low-cost bearing degradation detection by avoiding environmental interference, enabling timely preventive maintenance.
Smart Images

Figure CN119084246A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of generator detection, and in particular to a bearing degradation detection method, device, equipment and storage medium for a wind turbine generator. Background Art
[0002] Doubly-fed wind turbines are one of the most widely used power generation equipment. The main bearings inside them use locating ends and / or floating ends. When the generator set is running, the main bearings will generate a lot of heat under the action of their radial loads and axial loads. If the heat changes abnormally, it is more likely to deteriorate.
[0003] In order to monitor and detect the deterioration of bearings, the current main method is to add sensors at the monitored bearing positions to collect vibration, noise and other signals on the bearings to achieve monitoring and detection. However, this method is costly, and vibration, noise and other signals are easily disturbed by the surrounding environment, resulting in low detection accuracy. Summary of the invention
[0004] The main purpose of the present disclosure is to solve the problem that the existing generator bearing degradation detection scheme is easily affected by the environment, resulting in low detection accuracy.
[0005] A first aspect of the present disclosure provides a method for detecting bearing degradation of a wind turbine, comprising: collecting wind field data of the wind turbine at a current moment, the load acting on the bearing of the wind turbine, and the actual temperature of the bearing; calculating the real-time load of the bearing according to the wind field data and the load acting on the bearing; using a preset bearing temperature prediction model to predict the theoretical temperature generated by the bearing under the action of the real-time load; and judging whether the bearing has degraded according to the actual temperature and the theoretical temperature.
[0006] Optionally, the collecting of the wind field data of the wind turbine at the current moment, the load acting on the bearings of the wind turbine and the actual temperature of the bearings includes: collecting the vector wind speed and the vector linear velocity at a preset position in front of the blades of the wind turbine at the current moment, and calculating the vector inflow wind speed at the preset position based on the vector wind speed and the vector linear velocity; when the bearings are in a stationary state, respectively measuring the weight of the blades, the hub and the bearings of the wind turbine to obtain the load acting on the bearings; and measuring the actual temperature of the bearings of the wind turbine at the current moment by means of a temperature sensor.
[0007] Optionally, calculating the real-time load of the bearing based on the wind field data and the load acting on the bearing includes: calculating the axial load of the wind turbine based on the vector inflow wind speed, and calculating the real-time load of the bearing based on the axial load; or calculating the axial load of the wind turbine based on the vector inflow wind speed, determining the radial load of the wind turbine based on the load acting on the bearing, and calculating the real-time load of the bearing based on the axial load and the radial load.
[0008] Optionally, calculating the axial load of the wind turbine based on the vector inflow wind speed includes: calculating the installation angle of the blade based on blade design information in the wind turbine; decomposing the vector inflow wind speed to obtain inflow wind speed and inflow angle; and calculating the axial load of the wind turbine based on the installation angle, the inflow wind speed and the inflow angle.
[0009] Optionally, the calculating the axial load of the wind turbine based on the installation angle, the inflow wind speed and the inflow angle includes: determining the angle of attack of the wind force at a current moment relative to the blade based on the inflow angle and the installation angle; calculating the axial thrust coefficient of the bearing based on the angle of attack, the inflow angle and a coefficient variation curve of a preset force; and calculating the axial load of the wind turbine based on the axial thrust coefficient and the inflow wind speed.
[0010] Optionally, the method uses a preset bearing temperature prediction model to predict the theoretical temperature generated by the bearing under the action of the real-time load, including: inputting the real-time load into the preset bearing temperature prediction model, calculating the total heat generated by the bearing from an initial moment to a current moment; and predicting the theoretical temperature of the bearing at the current moment based on the total heat.
[0011] Optionally, judging whether the bearing has deteriorated based on the actual temperature and the theoretical temperature includes: calculating the temperature difference between the actual temperature and the theoretical temperature; judging whether the temperature difference is greater than a temperature difference threshold; if so, determining that the bearing has deteriorated and generating a bearing degradation warning; if not, continuing to detect the temperature change of the bearing.
[0012] Optionally, the method also includes: obtaining the temperature of the bearing of the wind turbine at each moment during a period of normal operation without degradation, and the axial load at each moment; based on the axial load at each moment and the radial load of the bearing, calculating the heat generated by the bearing from the initial moment to each moment; using the heat at each moment as an input parameter and the temperature at each moment as an output parameter, establishing and training a machine learning model to obtain a bearing temperature prediction model.
[0013] The second aspect of the present disclosure provides a bearing degradation detection device for a wind turbine, comprising: an acquisition module, used to acquire wind field data of the wind turbine at a current moment, the load acting on the bearing of the wind turbine and the actual temperature of the bearing; a calculation module, used to calculate the real-time load of the bearing according to the wind field data and the load acting on the bearing; a prediction module, used to predict the theoretical temperature generated by the bearing under the action of the real-time load by using a preset bearing temperature prediction model; and a detection module, used to determine whether the bearing has deteriorated according to the actual temperature and the theoretical temperature.
[0014] Optionally, the acquisition module includes: an acquisition unit, used to collect the vector wind speed and vector linear velocity at a preset position in front of the blades of the wind turbine at the current moment, and calculate the vector inflow wind speed at the preset position based on the vector wind speed and the vector linear velocity; a measuring unit, used to measure the weight of the blades, hub and bearings of the wind turbine respectively when the bearing is in a stationary state to obtain the load acting on the bearing; and measure the actual temperature of the bearing of the wind turbine at the current moment through a temperature sensor.
[0015] Optionally, the calculation module is specifically used to: calculate the axial load of the wind turbine based on the vector inflow wind speed, and calculate the real-time load of the bearing based on the axial load; or, calculate the axial load of the wind turbine based on the vector inflow wind speed, and determine the radial load of the wind turbine based on the load acting on the bearing; calculate the real-time load of the bearing based on the axial load and the radial load.
[0016] Optionally, the calculation module is specifically used to: calculate the installation angle of the blade based on the blade design information in the wind turbine; decompose the vector inflow wind speed to obtain the inflow wind speed and the inflow angle; calculate the axial load of the wind turbine based on the installation angle, the inflow wind speed and the inflow angle.
[0017] Optionally, the calculation module is specifically used to: determine the angle of attack of the wind force at the current moment relative to the blade based on the inflow angle and the installation angle; calculate the axial thrust coefficient of the bearing based on the coefficient change curve of the attack angle, the inflow angle and a preset force; calculate the axial load of the wind turbine based on the axial thrust coefficient and the inflow wind speed.
[0018] Optionally, the prediction module includes: a third calculation unit, used to input the real-time load into a preset bearing temperature prediction model to calculate the total heat generated by the bearing from the initial moment to the current moment; and a prediction unit, used to predict the theoretical temperature of the bearing at the current moment based on the total heat.
[0019] Optionally, the detection module includes: a fourth calculation unit, used to calculate the temperature difference between the actual temperature and the theoretical temperature; a judgment unit, used to judge whether the temperature difference is greater than a temperature difference threshold; an early warning unit, used to determine that the bearing has deteriorated and generate a bearing degradation warning when it is judged that the temperature difference is greater than the temperature difference threshold; and when it is judged that the temperature difference is not greater than the temperature difference threshold, continue to detect the temperature change of the bearing.
[0020] Optionally, the bearing degradation detection device for the wind turbine further includes: a training module, used to: obtain the temperature of the bearing of the wind turbine at each moment during a period of normal operation without degradation, and the axial load at each moment; based on the axial load at each moment and the radial load of the bearing, calculate the heat generated by the bearing from the initial moment to each moment; use the heat at each moment as an input parameter and the temperature at each moment as an output parameter, establish and train a machine learning model, and obtain a bearing temperature prediction model.
[0021] The third aspect of the present disclosure provides an electronic device, comprising: a memory and at least one processor, wherein instructions are stored in the memory, and the memory and the at least one processor are interconnected via lines; the at least one processor calls the instructions in the memory so that the electronic device executes the above-mentioned method for detecting bearing degradation of a wind turbine.
[0022] A fourth aspect of the present disclosure provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned method for detecting bearing degradation of a wind turbine.
[0023] Beneficial effects:
[0024] By implementing the technical solution provided by the present disclosure, the wind field data of the wind turbine at the current moment, the load of the unit acting on the bearing, and the actual temperature of the bearing of the wind turbine are collected; the real-time load of the bearing is calculated based on the wind field data and the load acting on the bearing; the theoretical temperature generated by the bearing under the action of the real-time load is predicted using a preset bearing temperature prediction model; and whether the bearing has deteriorated is determined based on the actual temperature and the theoretical temperature. By calculating the load generated by the wind and the generator itself on the bearing in real time, the theoretical temperature of the bearing is predicted based on the load and the bearing temperature prediction model, and the bearing is compared with the actual measured temperature to determine whether the bearing has deteriorated. This method can be implemented without additional testing, avoiding the problem of hardware being affected by the environment and affecting the test results, and the cost is also low. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 A first flow chart of a method for detecting bearing degradation of a wind turbine provided by an embodiment of the present disclosure;
[0026] Figure 2 A second flow chart of a method for detecting bearing degradation of a wind turbine provided by an embodiment of the present disclosure;
[0027] Figure 3 A third flow chart of a method for detecting bearing degradation of a wind turbine provided by an embodiment of the present disclosure;
[0028] Figure 4 A schematic diagram of the inflow angle, pitch angle and attack angle of a wind turbine provided in an embodiment of the present disclosure;
[0029] Figure 5 A schematic diagram of a support for a bearing of a wind turbine provided in an embodiment of the present disclosure;
[0030] Figure 6 Another support schematic diagram of a bearing of a wind turbine provided by an embodiment of the present disclosure;
[0031] Figure 7 A schematic diagram of the structure of a bearing degradation detection device for a wind turbine provided by an embodiment of the present disclosure;
[0032] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION
[0033] The disclosed embodiments provide a bearing degradation detection method, device, equipment and storage medium for a wind turbine. The specific technical solution is to calculate the axial thrust on the wind turbine in real time and establish a machine learning model of bearing load and bearing temperature. This can timely detect the bearing degradation trend, guide operation and maintenance personnel to perform preventive maintenance, and reduce the loss of major component replacement caused by bearing failure. No new sensors are required, and the temperature measurement points are not easily disturbed by the surrounding environment. It has the advantages of low cost and high accuracy.
[0034] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0035] The wind turbine in this application refers to a generator set structure with a two-point support bearing for the main shaft, and the main shaft refers to the shaft between the hub and the gearbox. Specifically, the generator set is supported by a positioning end and a floating end arrangement, that is, the bearing connected to the wind wheel is supported by two bearing seats, and the bearing seat close to the hub is used as a fixed end to withstand the thrust of the wind wheel; the bearing seat close to the gearbox is used as a floating end, so that the bearing can move axially when the temperature changes and the length changes, avoiding excessive expansion and contraction stress in the structure. The torque arm of the gearbox serves as an auxiliary support, or is connected to the frame through a pin elastic sleeve, or is connected to the cabin base through an elastic pad. A wind measuring radar is installed in front of the wind turbine set, which can be one or more. Of course, it can also be a single-point supported wind turbine. For the single-point support, the point of action of the bearing that simultaneously bears the radial load and the axial load is located at the same position, and the two-point radial load and the axial load are not at the same position.
[0036] Based on the above structural framework, the embodiment of the present application provides a method for detecting bearing degradation of a wind turbine generator, and the implementation process is as follows: Figure 1 As shown, the method comprises the following steps:
[0037] 101. Collect the wind field data of the wind turbine at the current moment, the load acting on the bearings of the wind turbine, and the actual temperature of the bearings.
[0038] In this embodiment, multiple sensors are set on the wind turbine to measure wind field data. The wind field data is actually vector inflow wind speed. The vector inflow wind speed is the relative speed of wind with respect to the blades and is calculated by measuring the speed and direction of the wind.
[0039] The load acting on the bearing of the wind turbine is actually the total weight of the blades, hub and main bearing on the generator set, which can be measured at the position of the bearing seat of the bearing by a pressure sensor after the generator is assembled or when the generator is static.
[0040] In actual applications, corresponding sensors are set up to use communication connections to collect and record the wind field data and temperature information of the bearing in real time, and store them in a storage device in chronological order. When the bearing needs to be deteriorated, the wind field data and temperature information at any time point or time period can be retrieved from the storage device for subsequent degradation detection.
[0041] 102. Calculate the real-time load of the bearing based on the wind field data and the load acting on the bearing.
[0042] Specifically, the wind field data and the load acting on the bearing are converted into corresponding forces respectively using a force conversion algorithm, and the two forces are summed to obtain a total force to obtain the real-time load of the bearing.
[0043] In this embodiment, since the wind field data is mainly calculated based on the wind speed, the vector wind speed at any position in front of the wind turbine blades is measured and recorded in real time by a wind measuring radar, and the vector inflow wind speed is obtained based on the vector wind speed and the wind turbine speed.
[0044] In practical applications, when converting wind field data into force, the axial and radial forces of wind on the bearing are calculated based on the wind field data, and the total force of the bearing is determined based on the load, axial and radial forces acting on the bearing, thereby obtaining the real-time load.
[0045] 103. Use the preset bearing temperature prediction model to predict the theoretical temperature of the bearing under the action of real-time load.
[0046] The bearing temperature prediction model is pre-trained based on the temperature rise data of non-degraded bearings under different loads. The real-time load obtained based on the wind field data and the load acting on the bearing is used as the input of the bearing temperature prediction model. After the model is calculated, the temperature of the bearing without degradation is obtained, which is the theoretical temperature.
[0047] Specifically, the bearing temperature prediction model is used to calculate the net heat generated by the bearing under the action of the real-time load. The net heat is the difference between the heat generated by the real-time load and the heat dissipated by the bearing, and then the net heat is converted into temperature.
[0048] 104. Determine whether the bearing has deteriorated based on the actual temperature and theoretical temperature.
[0049] The actual temperature is compared with the theoretical temperature. When the actual temperature is greater than the theoretical temperature, it is determined that the bearing has deteriorated. Otherwise, no deterioration has occurred.
[0050] In practical applications, since not all the heat generated is reflected in the temperature, there will be certain losses or errors. Therefore, when judging whether the bearing has deteriorated, the difference between the two can be calculated and then compared to see if the difference is within the allowable range to determine whether the bearing has deteriorated.
[0051] Furthermore, it is also possible to determine whether the bearing is deteriorated by judging whether the absolute value of the difference is within an allowable range.
[0052] In another embodiment, when judging bearing degradation, the degradation can be divided into multiple levels, and the default is to compare the judgment difference with the lower limit of degradation. When the lower limit is met, the error level is successively increased and the comparison is continued to determine the final degradation level.
[0053] In this embodiment, the wind field data of the wind turbine at the current moment, the load acting on the bearing of the wind turbine and the actual temperature of the bearing are collected; the real-time load of the bearing is calculated according to the wind field data and the load acting on the bearing; the theoretical temperature generated by the bearing under the action of the real-time load is predicted using a preset bearing temperature prediction model; and whether the bearing is deteriorated is determined according to the actual temperature and the theoretical temperature. This solves the problem that the existing generator bearing degradation detection scheme is easily affected by the environment, resulting in a low detection accuracy.
[0054] See also Figure 2 and 3 , which is another embodiment of the bearing degradation detection method for a wind turbine provided in the present application, comprises the following steps:
[0055] 201. Collect the vector wind speed and the vector linear velocity at a preset position in front of the blade of the wind turbine at the current moment, and calculate the vector inflow wind speed at the preset position based on the vector wind speed and the vector linear velocity.
[0056] The wind speed and wind direction at the corresponding position are collected by the speed sensor set at the preset position in front of the blade of the wind turbine, and the vector wind speed is obtained. Here, the wind direction from the fan blades is defined as the front, and the preset front position is the position where a wind measuring radar is installed on the wind side of the fan blades. This position can be set manually or based on experimental experience. For example, the wind measuring radar is usually installed at the front end of the fan shaft.
[0057] The same principle is used to collect the speed of the wind turbine, the distance from the preset position in front of the blade to the rotation center, and the azimuth of the blade rotation plane. The vector linear speed at the preset position is calculated based on the speed, distance and plane azimuth. It should be noted that the vector wind speed and vector line speed The information is at the same location.
[0058] Based on the vector wind speed and vector line speed Calculate the vector inflow wind speed at any position in front of the blade The vector inflow wind speed includes the inflow wind speed c and the inflow angle Specifically, the vector wind speed minus the vector line speed gives the vector inflow wind speed. That is
[0059] 202. Measure the actual temperature of the bearing of the wind turbine generator at the current moment through a temperature sensor.
[0060] 203. Calculate the axial load of the wind turbine based on the vector inflow wind speed.
[0061] Specifically, the installation parameters of the blades in the wind turbine are obtained, and the axial load of the wind turbine is calculated based on the installation parameters and the vector inflow wind speed. The installation parameters are calculated based on the design drawings of the wind turbine, and include the installation angle and the coefficient change curve of the force.
[0062] In one embodiment, the installation angle of the blade is calculated based on the blade design information in the wind turbine; the vector inflow wind speed is decomposed to obtain the inflow wind speed and the inflow angle; the angle of attack of the wind force at the current moment relative to the blade is determined based on the inflow angle and the installation angle; and the axial load of the wind turbine is calculated based on the angle of attack and the inflow wind speed.
[0063] Specifically, Figure 4 As shown, according to the design drawings of the blade, the blade is calculated under the condition that the pitch angle is zero, as well as the chord length at each position of the blade span and the torsion angle θ of the rotation plane; then the installation angle at each position of the blade span is calculated based on the pitch angle and the torsion angle θ, and the installation angle β=pitch angle+twist angle θ.
[0064] By solving the vector inflow wind speed, the inflow angle is extracted, and the angle of attack α of the blade at any position in the span direction is calculated based on the inflow angle and the installation angle.
[0065] Furthermore, the corresponding force coefficient is matched from the force coefficient variation curve based on the attack angle, and the axial thrust coefficient of the bearing is calculated based on the matched force coefficient and the inflow angle, and the axial thrust is further calculated, and the axial thrust is the axial load.
[0066] Specifically, the coefficients of the force include the lift coefficient and the drag coefficient. The lift coefficient C corresponding to the currently calculated attack angle α can be found from the curve of the lift coefficient and the drag coefficient changing with the attack angle. A and the drag coefficient C W , where the curves of lift coefficient and drag coefficient changing with angle of attack (i.e., the force coefficient change curve) are obtained by conducting wind tunnel tests on wind turbines.
[0067] Calculate the sine and cosine values of the inflow angle, and multiply the sine and cosine values by the lift coefficient and the drag coefficient to obtain the axial thrust coefficient C. T , the specific calculation formula is:
[0068] Finally, the axial thrust is obtained based on the axial thrust coefficient and the inflow wind speed, and the calculation formula is:
[0069] Where ρ is the air density, c is the inflow wind speed, x is the chord length at the point where the wind acts on the blade, and r is the distance from the point where the wind acts on the blade to the center of rotation of the blade.
[0070] 204. Calculate the real-time load of the bearing based on the axial load.
[0071] In this step, the bearing refers to a bearing based on two-point support. The point of action of the axial load and the point of action of the radial load are not at the same position and there is a certain distance between them. The real-time load of the bearing is mainly the axial load, such as Figure 5 As shown, the blades of the wind turbine are connected to a bearing 5A, and a fixed end 5B is provided at one end of the bearing 5A close to the blade, and the fixed end 5B is the point of application of the axial load, and a floating end 5C is provided at one end of the bearing 5A farther from the blade, and the floating end 5C is the point of application of the radial load.
[0072] In another embodiment, if the bearing adopts a single-point support, the real-time load of the bearing needs to also include a radial load. The structure diagram of the wind turbine with a single-point support is as follows: Figure 6 As shown, it is only provided with a fixed end 5A. Specifically, the radial load of the wind turbine is determined based on the weight of the wind turbine itself, wherein the weight of the wind turbine itself should be understood as the sum of the weights of the wind turbine blades, the hub and the bearings. The radial load can be obtained by measuring the force acting on the fixed end 5A of the bearing when the wind turbine main shaft is stationary.
[0073] When calculating the real-time load of bearing 5A, first determine the force vectors of the radial load and the axial load respectively, and then sum the two force vectors to obtain the real-time load of the bearing, such as Figure 3 shown.
[0074] 205. Use the preset bearing temperature prediction model to predict the theoretical temperature of the bearing under the action of real-time load.
[0075] Specifically, the real-time load is input into a preset bearing temperature prediction model to calculate the total heat generated by the bearing from an initial moment to a current moment; and the theoretical temperature of the bearing at the current moment is predicted based on the net heat.
[0076] In actual applications, when the fan rotates, the temperature of the bearing rises mainly because the radial load and axial load act on the inner ring of the bearing, and the relative displacement between the inner ring and the outer ring of the bearing generates heat, which causes the temperature of the bearing to rise. However, the bearing itself will also dissipate heat, and the net heat is obtained by subtracting the heat dissipated by the bearing to the outside world from the heat generated by the bearing, which causes the temperature of the bearing to rise.
[0077] The heat generated by the bearing is related to the load, the relative displacement of the inner and outer rings, and the lubrication performance. The heat dissipated by the bearing to the outside is related to the difference between the bearing temperature and the cabin temperature, that is, the heat dissipated increases with the increase of the temperature difference. According to the above relationship, the calculation formula of the net heat is constructed: ∫ radial load × time × bearing speed × lubrication performance + ∫ axial load × time × bearing speed × lubrication performance - ∫ (bearing temperature - cabin temperature) × heat transfer coefficient × heat transfer area ÷ heat transfer thickness × time = bearing specific heat capacity × (bearing temperature - initial bearing temperature), where time × bearing speed reflects the relative displacement of the inner and outer rings of the bearing, and the initial bearing temperature is the bearing temperature at the initial moment manually confirmed, and the moment before the bearing rotates can be used as the initial moment; after calculating the net heat, the total temperature of the bearing raised by the net heat, that is, the theoretical temperature, is calculated based on the heat required for the bearing to rise in unit temperature (that is, the specific heat capacity) and the initial bearing temperature.
[0078] 206. Determine whether the bearing has deteriorated based on the actual temperature and theoretical temperature.
[0079] Specifically, by calculating the temperature difference between the actual temperature and the theoretical temperature; determining whether the temperature difference is greater than a temperature difference threshold; if so, determining that the bearing has deteriorated, and generating a bearing degradation warning; if not, continuing to detect the temperature change of the bearing.
[0080] The integrated value of (radial load × bearing speed + axial load × bearing speed) over time and the integrated value of (bearing temperature - engine room temperature) over time in the time period from the initial moment to the current moment are input into the trained model to obtain the model-predicted bearing temperature rise. The predicted bearing temperature at the current moment = initial bearing temperature + model-predicted bearing temperature rise. The predicted bearing temperature at the current moment is compared with the actual bearing temperature. When the actual bearing temperature - predicted bearing temperature > temperature difference threshold a (a is a threshold greater than 0), a bearing degradation warning is issued.
[0081] In this embodiment, the above-mentioned bearing temperature prediction model is obtained by training using a neural network combined with the temperature recording data of the bearing. Specifically: the temperature of the bearing of the wind turbine during normal operation for a period of time without degradation and the axial load at each moment are obtained; based on the axial load at each moment and the radial load of the bearing, the heat generated by the bearing from the initial moment to each moment is calculated; the heat at each moment is used as an input parameter and the temperature at each moment is used as an output parameter, a machine learning model is established and trained, and a bearing temperature prediction model is obtained.
[0082] For example, by selecting a period of normal operation of the bearing, starting from the initial moment during this period, the integral value of (radial load × bearing speed + axial load × bearing speed) over time and the integral value of (bearing temperature - engine room temperature) over time at each moment are used as input parameters, and the main bearing temperature - initial bearing temperature at each moment is used as the output parameter, a machine learning model is established and trained to obtain a bearing temperature prediction model.
[0083] In summary, the real-time axial thrust of the wind turbine is calculated based on the real-time wind field parameters measured by the wind radar, and the real-time load of the bearing is obtained, which solves the problem of direct measurement of the axial thrust during the operation of the unit; further, by establishing a machine learning model between the bearing load and the bearing temperature, it is possible to timely warn of the bearing degradation trend without the need for new sensors, and the temperature is not easily affected by the surrounding environment, which has the advantages of low cost and high accuracy.
[0084] The above describes the bearing degradation detection method of the wind turbine in the embodiment of the present disclosure. The following describes the bearing degradation detection device of the wind turbine in the embodiment of the present disclosure. Figure 7 , an embodiment of a bearing degradation detection device for a wind turbine in the embodiment of the present disclosure includes:
[0085] The acquisition module 710 is used to acquire the wind field data of the wind turbine at the current moment, the load acting on the bearing of the wind turbine and the actual temperature of the bearing;
[0086] A calculation module 720, configured to calculate a real-time load of the bearing according to the wind field data and the load acting on the bearing;
[0087] A prediction module 730, configured to predict the theoretical temperature of the bearing under the real-time load using a preset bearing temperature prediction model;
[0088] The detection module 740 is used to determine whether the bearing is deteriorated according to the actual temperature and the theoretical temperature.
[0089] Optionally, the acquisition module 710 includes:
[0090] The collecting unit 7101 is used to collect the vector wind speed and the vector linear speed of the preset position in front of the blade of the wind turbine at the current moment, and calculate the vector inflow wind speed of the preset position based on the vector wind speed and the vector linear speed;
[0091] The measuring unit 7102 is used to measure the weight of the blades, hub and bearing of the wind turbine respectively when the bearing is in a stationary state to obtain the load acting on the bearing; and to measure the actual temperature of the bearing of the wind turbine at the current moment through a temperature sensor.
[0092] Optionally, the calculation module 720 is specifically used to: calculate the axial load of the wind turbine based on the vector inflow wind speed, and calculate the real-time load of the bearing based on the axial load; or, calculate the axial load of the wind turbine based on the vector inflow wind speed, and determine the radial load of the wind turbine based on the load acting on the bearing; calculate the real-time load of the bearing based on the axial load and the radial load.
[0093] Optionally, the calculation module 720 is specifically used for:
[0094] The installation angle of the blade is calculated based on the blade design information in the wind turbine; the vector inflow wind speed is decomposed to obtain the inflow wind speed and the inflow angle; and the axial load of the wind turbine is calculated based on the installation angle, the inflow wind speed and the inflow angle.
[0095] Optionally, the calculation module 720 is specifically used to: determine the angle of attack of the wind force at the current moment relative to the blade based on the inflow angle and the installation angle; calculate the axial thrust coefficient of the bearing based on the coefficient change curve of the angle of attack, the inflow angle and a preset force; calculate the axial load of the wind turbine based on the axial thrust coefficient and the inflow wind speed.
[0096] Optionally, the prediction module 730 includes:
[0097] The first calculation unit 7301 is used to input the real-time load into a preset bearing temperature prediction model to calculate the total heat generated by the bearing from an initial moment to a current moment;
[0098] The prediction unit 7302 is used to predict the theoretical temperature of the bearing at the current moment based on the total heat.
[0099] Optionally, the detection module 740 includes:
[0100] The second calculation unit 7401 is used to calculate the temperature difference between the actual temperature and the theoretical temperature;
[0101] A judging unit 7402 is used to judge whether the temperature difference is greater than a temperature difference threshold;
[0102] The early warning unit 7403 is used to determine that the bearing has deteriorated and generate a bearing degradation early warning when it is determined that the temperature difference is greater than the temperature difference threshold; and to continue to detect the temperature change of the bearing when it is determined that the temperature difference is not greater than the temperature difference threshold.
[0103] Optionally, the bearing degradation detection device for a wind turbine further includes: a training module 750, which is used to:
[0104] Obtain the temperature of the wind turbine bearing at each moment during normal operation for a period of time without degradation, and the axial load at each moment;
[0105] Based on the axial load at each moment and the radial load of the bearing, the heat generated by the bearing from the initial moment to each moment is calculated;
[0106] The heat at each moment is used as the input parameter, and the temperature at each moment is used as the output parameter. A machine learning model is established and trained to obtain a bearing temperature prediction model.
[0107] Through the implementation of this embodiment, the real-time load on the wind turbine is calculated based on the measured wind field data; the real-time temperature of the bearing is predicted based on the real-time load and the bearing temperature prediction model established in advance using the load and the temperature of the bearing, and when it is determined that the actual bearing temperature is significantly greater than the real-time temperature predicted by the model, a bearing degradation warning is issued.
[0108] above Figure 7 The bearing degradation detection device for a wind turbine in an embodiment of the present disclosure is described in detail from the perspective of modular functional entities. The bearing degradation detection device for a wind turbine in an embodiment of the present disclosure is described in detail from the perspective of hardware processing.
[0109] Figure 8 8 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. The electronic device 800 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 810 (for example, one or more processors), a memory 820, a storage medium 830 and a camera 840. A series of instruction operations in the storage medium 830 are executed on the electronic device 800. Those skilled in the art can understand that Figure 8 The structure of the electronic device shown does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.
[0110] The present disclosure also provides an electronic device, which includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the bearing degradation detection method for the wind turbine provided in the above embodiments.
[0111] The present disclosure also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein instructions are stored in the computer, and when the instructions are executed on a computer, the computer executes the method for detecting bearing degradation of a wind turbine provided in the above-mentioned embodiment.
[0112] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.
[0113] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.
[0114] As described above, the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure.
Claims
1. A method for detecting bearing degradation of a wind turbine, characterized in that: The method comprises: Collecting wind field data of the wind turbine at the current moment, load acting on the bearing of the wind turbine, and actual temperature of the bearing; Calculating the real-time load of the bearing according to the wind field data and the load acting on the bearing; Using a preset bearing temperature prediction model, predicting the theoretical temperature generated by the bearing under the action of the real-time load; Whether the bearing is deteriorated is determined based on the actual temperature and the theoretical temperature.
2. The method for detecting bearing degradation of a wind turbine according to claim 1, characterized in that: The collecting of wind field data of the wind turbine at the current moment, load acting on the bearing of the wind turbine and actual temperature of the bearing includes: Collecting the vector wind speed and the vector linear velocity at a preset position in front of the blade of the wind turbine at the current moment, and calculating the vector inflow wind speed at the preset position based on the vector wind speed and the vector linear velocity; When the bearing is in a stationary state, the weight of the blades, the hub and the bearing of the wind turbine are measured respectively to obtain the load acting on the bearing; The actual temperature of the bearing of the wind turbine generator at the current moment is measured by a temperature sensor.
3. The method for detecting bearing degradation of a wind turbine according to claim 2, characterized in that: The calculating the real-time load of the bearing according to the wind field data and the load acting on the bearing comprises: Calculating the axial load of the wind turbine based on the vector inflow wind speed, and calculating the real-time load of the bearing based on the axial load; or, The axial load of the wind turbine is calculated based on the vector inflow wind speed, and the radial load of the wind turbine is determined based on the load acting on the bearing; the real-time load of the bearing is calculated based on the axial load and the radial load.
4. The method for detecting bearing degradation of a wind turbine according to claim 3, characterized in that: The step of calculating the axial load of the wind turbine generator based on the vector inflow wind speed comprises: Calculating the installation angle of the blade based on the blade design information of the wind turbine; Decomposing the vector inflow wind speed to obtain inflow wind speed and inflow angle; The axial load of the wind turbine is calculated based on the installation angle, the inflow wind speed and the inflow angle.
5. The method for detecting bearing degradation of a wind turbine according to claim 4, characterized in that: The step of calculating the axial load of the wind turbine generator based on the installation angle, the inflow wind speed and the inflow angle comprises: Determine the angle of attack of the wind force at the current moment relative to the blade based on the inflow angle and the installation angle; Calculating the axial thrust coefficient of the bearing based on the angle of attack, the inflow angle and a coefficient variation curve of a preset force; The axial load of the wind turbine is calculated based on the axial thrust coefficient and the inflow wind speed.
6. The method for detecting bearing degradation of a wind turbine according to any one of claims 1 to 5, characterized in that: The method of using a preset bearing temperature prediction model to predict the theoretical temperature of the bearing under the action of the real-time load includes: Inputting the real-time load into a preset bearing temperature prediction model to calculate the total heat generated by the bearing from the initial moment to the current moment; The theoretical temperature of the bearing at the current moment is predicted based on the total heat.
7. The method for detecting bearing degradation of a wind turbine according to any one of claims 1 to 5, characterized in that: The determining whether the bearing is deteriorated according to the actual temperature and the theoretical temperature includes: Calculating the temperature difference between the actual temperature and the theoretical temperature; Determining whether the temperature difference is greater than a temperature difference threshold; If so, it is determined that the bearing has deteriorated, and a bearing degradation warning is generated; If not, continue to detect the temperature change of the bearing.
8. The method for detecting bearing degradation of a wind turbine according to claim 1, characterized in that: The method further comprises: Obtain the temperature of the wind turbine bearing at each moment during normal operation for a period of time without degradation, and the axial load at each moment; Based on the axial load at each moment and the radial load of the bearing, the heat generated by the bearing from the initial moment to each moment is calculated; The heat at each moment is taken as the input parameter, and the temperature at each moment is taken as the output parameter. A machine learning model is established and trained to obtain a bearing temperature prediction model.
9. A bearing degradation detection device for a wind turbine, characterized in that: The device comprises: A collection module, used to collect wind field data of the wind turbine at the current moment, loads acting on the bearings of the wind turbine, and actual temperatures of the bearings; A calculation module, used for calculating the real-time load of the bearing according to the wind field data and the load acting on the bearing; A prediction module, used to predict the theoretical temperature generated by the bearing under the action of the real-time load by using a preset bearing temperature prediction model; The detection module is used to determine whether the bearing is deteriorated according to the actual temperature and the theoretical temperature.
10. An electronic device, characterized in that: The electronic device includes a processor and a memory, the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the bearing degradation detection method for a wind turbine according to any one of claims 1-8.
11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for detecting bearing degradation of a wind turbine according to any one of claims 1 to 8 is implemented.
