Method, device and wind turbine for detecting abnormal bearing temperature of wind turbine

Abstract

The present invention provides a method and device for detecting abnormal bearing temperature of a wind turbine and a wind power unit, including: obtaining a monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under near-full-load operating conditions, and obtaining the bearing temperature and ambient temperature at each sampling moment therefrom; determining a bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculating the temperature difference value between the bearing temperature and the bearing reference temperature; determining the temperature state of the bearing according to the temperature difference values at multiple consecutive sampling moments. According to the temperature difference value between the bearing temperature and the bearing temperature trend threshold line at each current sampling moment, the present invention can accurately judge the bearing temperature rise of the wind turbine, use it as a reference standard for judging the bearing temperature state, can deeply utilize the bearing temperature rise law of the wind turbine, improves the accuracy and credibility of the abnormal bearing temperature detection result, and has a longer early warning lead time.

Description

Technical Field

[0001] The present invention relates to the technical field of new power energy, and particularly relates to a method and device for detecting abnormal bearing temperature of a wind turbine and a wind turbine unit. Background Art

[0002] Bearings are the core components of a wind power generation unit (hereinafter referred to as a wind turbine). Once a bearing is damaged, it needs to be replaced by lowering the tower. The hoisting and maintenance costs are very high and the spare part cycle is long. Therefore, it is very necessary to establish an early warning model for bearing abnormality detection.

[0003] In the prior art, a neural network is generally used to establish a temperature prediction model, and the bearing temperature detected in real time is compared with the output value of the temperature prediction model to determine whether the bearing temperature is abnormal.

[0004] Since the training process of the temperature prediction model is relatively complex, and the temperature prediction model outputs a theoretical normal value, directly judging the bearing temperature by the temperature prediction model may have a risk of misjudgment. Summary of the Invention

[0005] The present invention provides a method and device for detecting abnormal bearing temperature of a wind turbine and a wind turbine unit, so as to solve the defect that the existing technology generally uses a neural network for bearing temperature abnormality detection and is inaccurate.

[0006] In a first aspect, the present invention provides a method for detecting abnormal bearing temperature of a wind turbine, including:

[0007] Obtain a monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under a near-full-load condition. The monitoring data set includes the bearing temperature and the ambient temperature collected at each sampling moment;

[0008] Screen out the bearing temperature and the ambient temperature collected at each sampling moment of the wind turbine to be measured within a first preset time period from all the monitoring data;

[0009] For the bearing temperature and the ambient temperature collected at any sampling moment, determine a corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculate the temperature difference value between the bearing temperature and the bearing reference temperature;

[0010] Determine the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset time period.

[0011] According to the method for detecting abnormal bearing temperature of a wind turbine provided by the present invention, the determining the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset time period includes:

[0012] Obtain the average value of the temperature difference values of the to-be-tested wind turbine at all sampling moments within the current preset sub-duration within the first preset duration as the first temperature difference average value;

[0013] Obtain the average value of the temperature difference values of the to-be-tested wind turbine at all sampling moments within the historical preset sub-duration within the first preset duration as the second temperature difference average value;

[0014] If the first temperature difference average value is greater than the second temperature difference average value, it is determined that the bearing temperature rise of the to-be-tested wind turbine within the first preset duration is positive, and then it is determined that the first bearing temperature state of the to-be-tested wind turbine is abnormal bearing temperature.

[0015] According to a method for detecting abnormal bearing temperature of a wind turbine provided by the present invention, when it is determined that the bearing temperature rise of the to-be-tested wind turbine within the first preset duration is positive, it further includes:

[0016] If the difference between the first temperature difference average value and the second temperature difference average value is greater than a third temperature threshold, it is determined that the bearing temperature state of the to-be-tested wind turbine is a serious bearing temperature fault;

[0017] If the difference between the first temperature difference average value and the second temperature difference average value is not greater than a first temperature threshold but greater than a second temperature threshold, it is determined that the first bearing temperature state of the to-be-tested wind turbine is a minor bearing temperature fault.

[0018] According to a method for detecting abnormal bearing temperature of a wind turbine provided by the present invention, the bearing temperature trend threshold line is pre-constructed in the following manner:

[0019] Obtain a first historical monitoring data set of all wind turbines in the wind farm under near-full-load operating conditions; the first historical monitoring data set includes the bearing temperature and ambient temperature of each wind turbine collected at multiple historical sampling moments;

[0020] Taking the ambient temperature and the bearing temperature as the X-axis and Y-axis respectively, construct a rectangular coordinate system, and mark the bearing temperature and ambient temperature collected at each historical sampling moment as a sample point in the rectangular coordinate system;

[0021] Perform linear fitting on all sample points to obtain the slope of the initially obtained threshold line as the slope of the bearing temperature trend threshold line;

[0022] Obtain a second historical monitoring data set of the to-be-tested wind turbine under near-full-load operating conditions within a recent second preset duration;

[0023] When the number of historical monitoring data in the second historical monitoring data set is greater than the first quantity threshold, the bearing temperature and the ambient temperature collected at each sampling moment in the second historical monitoring data set are used as a set of fitting data, and an expected intercept of the bearing temperature trend threshold line is calculated in combination with the slope of the bearing temperature trend threshold line;

[0024] The average value of the expected intercepts calculated for all sampling moments is used as the intercept of the bearing temperature trend threshold line.

[0025] According to a method for detecting abnormal bearing temperature of a wind turbine provided by the present invention, when the number of historical monitoring data in the second historical monitoring data set is not greater than the first quantity threshold, it further includes:

[0026] Obtain a third historical monitoring data set of other wind turbines of the same model as the wind turbine to be tested in the wind farm within the second preset duration under a near-full load condition;

[0027] The bearing temperature and the ambient temperature collected at each sampling moment in the third historical monitoring data set are used as a set of fitting data to determine the intercept of the bearing temperature trend threshold line.

[0028] According to a method for detecting abnormal bearing temperature of a wind turbine provided by the present invention, it further includes:

[0029] Screen out the bearing temperature collected at each sampling moment of the wind turbine to be tested within the third preset duration from all monitoring data;

[0030] According to the bearing temperature collected at each sampling moment, calculate the daily average bearing temperature of the wind turbine to be tested within the third preset duration;

[0031] If at least one of the daily average bearing temperatures is greater than the third temperature threshold and all daily average bearing temperatures are less than the fourth temperature threshold, it is determined that the second bearing temperature state of the wind turbine to be tested is a slight bearing temperature fault;

[0032] If there is at least one of the daily average bearing temperatures greater than or equal to the fourth temperature threshold, it is determined that the second bearing temperature state of the wind turbine to be tested is a serious bearing temperature fault; the fourth temperature threshold is greater than the third temperature threshold;

[0033] Determine the true bearing temperature state of the wind turbine to be tested according to the first bearing temperature state and the second bearing temperature state.

[0034] According to a method for detecting abnormal bearing temperature of a wind turbine provided by the present invention, it further includes:

[0035] Based on the monitoring data set of all wind turbines in the wind farm under near-full-load operating conditions, calculate the average bearing temperature and temperature standard deviation of all wind turbines in the wind farm, so as to determine the first allowable range and the second allowable range of the bearing temperature according to the average bearing temperature and temperature standard deviation; the second allowable range of the bearing temperature is greater than the first allowable range of the bearing temperature;

[0036] If the bearing temperature collected at any sampling moment of the wind turbine to be measured within the fourth preset duration exceeds the first allowable range of the bearing temperature but is within the second allowable range of the bearing temperature, determine that the third bearing temperature state is a slight bearing temperature fault;

[0037] If the bearing temperature collected at any sampling moment of the wind turbine to be measured within the fourth preset duration exceeds the second allowable range of the bearing temperature, determine the third bearing temperature state of the wind turbine to be measured;

[0038] According to the third bearing temperature state and the true bearing temperature state, determine the final bearing temperature state of the wind turbine to be measured.

[0039] According to a method for detecting abnormal bearing temperature of a wind turbine provided by the present invention, the determining the final bearing temperature state of the wind turbine to be measured according to the third bearing temperature state and the true bearing temperature state includes:

[0040] If any one of the results of the third bearing temperature state and the final bearing temperature state is a serious bearing temperature fault, determine that the final bearing temperature state of the wind turbine to be measured is a serious bearing temperature fault;

[0041] If neither of the results of the third bearing temperature state and the final bearing temperature state is a serious bearing temperature fault, but any one of the results is a slight bearing temperature fault, determine that the final bearing temperature state of the wind turbine to be measured is a slight bearing temperature fault;

[0042] Otherwise, determine that the final bearing temperature state of the wind turbine to be measured is normal.

[0043] In a second aspect, the present invention further provides a device for detecting abnormal bearing temperature of a wind turbine, including:

[0044] A data acquisition unit, configured to obtain a monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under near-full-load operating conditions, and the monitoring data set includes the bearing temperature and the ambient temperature collected at each sampling moment;

[0045] A data screening unit, configured to screen out the bearing temperature and the ambient temperature collected at each sampling moment of the wind turbine to be measured within the first preset duration from all the monitoring data;

[0046] A temperature analysis unit, configured to, for the bearing temperature and the ambient temperature collected at any sampling moment, determine a corresponding bearing reference temperature from a bearing temperature trend threshold line according to the ambient temperature, and calculate a temperature difference value between the bearing temperature and the bearing reference temperature;

[0047] A result prediction unit, configured to determine a first bearing temperature state of the to-be-detected wind turbine generator set according to the temperature difference values corresponding to each sampling moment within the first preset duration.

[0048] In a third aspect, the present invention provides a wind turbine generator set, including a wind turbine generator set body, in which a detection processor is provided; and further including a memory and a program or instruction stored on the memory and executable on the detection processor, when the program or instruction is executed by the detection processor, it implements the steps of the method for detecting abnormal bearing temperature of a wind turbine generator set as described in any one of the above. The wind turbine generator set includes a wind turbine generator set body, in which a detection processor is provided; and further includes a memory and a program or instruction stored on the memory and executable on the detection processor, when the program or instruction is executed by the detection processor, it implements the steps of the method for detecting abnormal bearing temperature of a wind turbine generator set as described in any one of the above.

[0049] In a fourth aspect, the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, when the processor executes the program, it implements the steps of the method for detecting abnormal bearing temperature of a wind turbine generator set as described in any one of the above.

[0050] In a fifth aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, when the computer program is executed by a processor, it implements the steps of the method for detecting abnormal bearing temperature of a wind turbine generator set as described in any one of the above.

[0051] The method, device and wind turbine generator set for detecting abnormal bearing temperature provided by the present invention can accurately judge the bearing temperature rise of the wind turbine generator set according to the temperature difference value between the bearing temperature and the bearing temperature trend threshold line at each current sampling moment, and use it as a reference standard for judging the bearing temperature state, can deeply utilize the bearing temperature rise law of the wind turbine generator set, improve the accuracy and credibility of the detection result of abnormal bearing temperature, and have a longer early warning time. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0053] Figure 1 It is one of the schematic flowcharts of the method for detecting abnormal bearing temperature of a wind turbine provided by the present invention;

[0054] Figure 2 It is the schematic flowchart of constructing the bearing temperature trend threshold line provided by the present invention;

[0055] Figure 3 It is the schematic diagram of a bearing temperature trend threshold line provided by the present invention;

[0056] Figure 4 It is the schematic diagram of the temperature difference between the bearing temperature of a wind turbine and the bearing temperature trend threshold line provided by the present invention;

[0057] Figure 5 It is the second schematic flowchart of the method for detecting abnormal bearing temperature of a wind turbine provided by the present invention;

[0058] Figure 6 It is the schematic structural diagram of the device for detecting abnormal bearing temperature of a wind turbine provided by the present invention;

[0059] Figure 7 It is the schematic structural diagram of the electronic device provided by the present invention. Detailed implementation manners

[0060] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without making creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0061] It should be noted that in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element. The orientation or positional relationship indicated by terms such as "upper", "lower", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation on the present invention. Unless otherwise clearly specified and defined, the terms "mount", "connect", "couple" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, and can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0062] The terms "first", "second", etc. in the description and claims of this application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances, so that the embodiments of this application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually of the same kind, and do not limit the number of objects. For example, the first object can be one or multiple. In addition, "and / or" in the description and claims means at least one of the connected objects, and the character " / ", generally represents an "or" relationship between the associated objects before and after.

[0063] It should be noted in advance that the method for detecting abnormal bearing temperature of a wind turbine provided by the present invention can be used to detect abnormal temperature of each bearing on the wind turbine, including but not limited to the main bearing of the generator (hereinafter referred to as the main bearing), yaw bearing, pitch bearing, gearbox bearing, etc. Without special explanation, the present invention takes the detection of abnormal temperature of the main bearing as an example for description, and the method for detecting abnormal bearing temperature of the wind turbine provided by the present invention can be migrated to the detection of abnormal temperature of other bearings. In the embodiment, the detection of abnormal temperature of the main bearing is taken as an example for description. Correspondingly, the ambient temperature refers to the temperature in the nacelle.

[0064] Figure 1It is one of the flow schematic diagrams of the method for detecting abnormal bearing temperature of a wind turbine provided by the present invention. As Figure 1 shown, it mainly includes but is not limited to the following steps:

[0065] Step 101: Obtain the monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under near-full-load operating conditions.

[0066] Among them, the monitoring data set includes the bearing temperature and ambient temperature collected at each sampling moment.

[0067] Optionally, retrieve the historical operation data of all wind turbines in the same wind farm as the wind turbine to be measured from the SCADA system, and filter and screen out the bearing temperature and ambient temperature collected at each sampling moment of each wind turbine under near-full-load operating conditions to construct a monitoring data set.

[0068] Among them, the monitoring data under any near-full-load operating condition refers to a set of monitoring data collected when the operating power of the wind turbine is close to the rated power. For example, it can be set that the bearing temperature and ambient temperature collected at any sampling moment when the operating power of the wind turbine is within the range of rated power*(1 - 10%) are used as a set of monitoring data.

[0069] The reason why the present invention limits the acquisition of the bearing temperature and ambient temperature of all wind turbines under near-full-load operating conditions is to take into account the close correlation between the bearing temperature and the ambient temperature. Only in this way, the corresponding inspection results obtained by horizontally comparing the bearing temperature of the wind turbine to be measured with the bearing temperatures of other wind turbines in the same wind farm under the same ambient temperature are meaningful.

[0070] The monitoring data set includes not only the bearing temperature and ambient temperature collected at each sampling moment (for example, collected every 5 minutes) of the wind turbine to be measured within a certain period of time (for example, two years), but also the bearing temperature and ambient temperature collected at each sampling moment of all other wind turbines in the same wind farm.

[0071] As an optional embodiment, the bearing temperature and ambient temperature of each wind turbine collected within a certain period of time can be preprocessed. For example, assuming that 12 sets of bearing temperature and ambient temperature are sampled within 1 hour, the average value of all bearing temperatures collected within 1 hour can be calculated as the bearing temperature collected per hour, and the average value of all ambient temperatures collected within 1 hour can be calculated as the ambient temperature collected per hour to construct a monitoring data set. The sampling period of each set of monitoring data in this monitoring data set is once per hour. In this way, the data volume of the actually collected monitoring data can be further reduced, and the average value method can also eliminate the influence of outliers on the actual detection results.

[0072] Furthermore, due to the possible occurrence of temperature data jumps, it is necessary to pre-filter all the monitoring data in the monitoring dataset to extract the monitoring data with adjacent temperature differences within the normal range and eliminate the null values.

[0073] Among them, eliminating null values means that if any one or more of the four variables, namely time, bearing temperature, ambient temperature, and operating power, in a set of monitoring data are null, then this set of monitoring data will be deleted from the monitoring dataset.

[0074] Step 102: Screen out the bearing temperature and ambient temperature collected at each sampling moment of the to-be-tested wind turbine within the first preset time period from all the monitoring data.

[0075] The present invention can judge the bearing temperature state according to the bearing temperature rise of the to-be-tested wind turbine within a certain time period.

[0076] Optionally, the present invention first screens out the bearing temperature and ambient temperature collected at each sampling moment of the to-be-tested wind turbine within the first preset time period (e.g., one week) from the monitoring dataset constructed in step 101, in order to analyze the bearing temperature rise of the to-be-tested wind turbine within the first preset time period through the screened detection data.

[0077] Step 103: For the bearing temperature and ambient temperature collected at any sampling moment, determine the corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculate the temperature difference value between the bearing temperature and the bearing reference temperature.

[0078] In order to quantify the bearing temperature rise of the to-be-tested wind turbine within the first preset time period, the present invention can pre-fit a bearing temperature trend threshold line that correlates the bearing temperature with the ambient temperature according to the historical operation data of all the wind turbines in the wind farm where the to-be-tested wind turbine is located (collected under normal operation conditions of each wind turbine).

[0079] In this way, for the ambient temperature collected at any sampling moment of the to-be-tested wind turbine within the first preset time period, the bearing temperature corresponding to it on the bearing temperature trend threshold line (referred to as: bearing reference temperature for easy distinction) can be determined according to the bearing temperature trend threshold line, so that the temperature difference value between the bearing reference temperature and the bearing temperature actually collected at this sampling moment can be compared.

[0080] Based on the magnitude of this temperature difference value, it can be further judged whether the bearing temperature collected at each sampling moment is abnormal compared to the bearing reference temperature under normal operation conditions.

[0081] Step 104: Determine the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset duration.

[0082] By calculating the temperature difference values between the bearing reference temperature related to each sampling moment within the first preset duration and the bearing temperature actually collected at that sampling moment, it is possible to determine the bearing temperature rise change trend of the wind turbine to be measured within the first preset duration according to the change of the temperature difference values at all sampling moments in time series. Furthermore, it is possible to determine whether the current bearing temperature state of the wind turbine to be measured is normal or abnormal based on the temperature rise change trend. (For the sake of description, the bearing temperature state obtained in this embodiment is referred to as the first bearing temperature state).

[0083] Generally speaking, if there is an upward trend in the temperature rise, it can be determined that the first bearing temperature state is: the bearing temperature is abnormal; correspondingly, if there is no upward trend in the temperature rise, it is determined that the first bearing temperature state is: the bearing temperature is normal.

[0084] The wind turbine bearing temperature anomaly detection method provided by the present invention can accurately judge the bearing temperature rise of the wind turbine according to the temperature difference values between the bearing temperature at each current sampling moment and the bearing temperature trend threshold line, and use it as a reference standard for judging the bearing temperature state. It can deeply utilize the bearing temperature rise law of the wind turbine, improve the accuracy and credibility of the bearing temperature anomaly detection result, and have a longer early warning time.

[0085] Based on the content of the above embodiment, as an alternative embodiment, the determining the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset duration includes:

[0086] Obtain the average value of the temperature difference values at all sampling moments of the wind turbine to be measured within the current preset sub-duration within the first preset duration as the first temperature difference average value;

[0087] Obtain the average value of the temperature difference values at all sampling moments of the wind turbine to be measured within the historical preset sub-duration within the first preset duration as the second temperature difference average value;

[0088] If the first temperature difference average value is greater than the second temperature difference average value, it is judged that the bearing temperature rise of the wind turbine to be measured within the first preset duration is positive, and it is determined that the first bearing temperature state of the wind turbine to be measured is that the bearing temperature is abnormal.

[0089] Figure 2 It is a schematic flow chart of constructing the bearing temperature trend threshold line provided by the present invention. As an alternative embodiment, as Figure 2As shown, the bearing temperature trend threshold line is pre-constructed in the following manner:

[0090] Obtain the first historical monitoring data set of all wind turbines in the wind farm under near-full-load conditions; the first historical monitoring data set includes the bearing temperatures and ambient temperatures of each wind turbine collected at multiple historical sampling times;

[0091] Taking the ambient temperature and bearing temperature as the X-axis and Y-axis respectively, construct a rectangular coordinate system, and mark the bearing temperature and ambient temperature collected at each historical sampling time as a sample point in the rectangular coordinate system;

[0092] Perform linear fitting on all sample points to obtain the slope of the initially obtained threshold line as the slope of the bearing temperature trend threshold line;

[0093] Obtain the second historical monitoring data set of the wind turbine to be measured under near-full-load conditions within a second preset time period recently;

[0094] When the number of historical monitoring data in the second historical monitoring data set is greater than the first quantity threshold, take the bearing temperature and ambient temperature collected at each sampling time in the second historical monitoring data set as a set of fitting data, and calculate an expected intercept of the bearing temperature trend threshold line in combination with the slope of the bearing temperature trend threshold line;

[0095] Take the average value of the expected intercepts calculated for all sampling times as the intercept of the bearing temperature trend threshold line.

[0096] Specifically, the method for constructing the bearing temperature trend threshold line provided by the present invention generally includes two major parts:

[0097] The first part is to determine the slope of the bearing temperature trend threshold line based on the first historical monitoring data set of all wind turbines in the wind farm under near-full-load conditions.

[0098] The second part is to determine the intercept of the bearing temperature trend threshold line according to the second historical monitoring data set of the wind turbine to be measured under near-full-load conditions recently.

[0099] Both the above-mentioned first historical monitoring data set and second historical monitoring data set can be determined by calling the Scada data of the wind turbine.

[0100] As an optional embodiment, it is possible to use the average bearing temperature and average nacelle temperature of all wind turbines in the wind farm where the wind turbine to be measured is located within each second preset time period (such as 1 day).

[0101] Mark a sample point in a rectangular coordinate system with the ambient temperature and the bearing temperature as the X-axis and Y-axis respectively, that is, the average bearing temperature and the average nacelle temperature of each wind turbine within each second preset time period correspond to a sample point in the rectangular coordinate system.

[0102] Then, perform linear fitting on all sample points to obtain a full-wind-field fitting line, use it as the initial threshold line, and obtain its slope as the slope of the bearing temperature trend threshold line.

[0103] As another alternative embodiment, first select all data of all wind farms for filtering under near-full-load conditions, and count the average bearing temperature and the average nacelle temperature of each wind farm within each preset time period (such as 1 day). Then, according to the distribution relationship between the average bearing temperature and the average nacelle temperature, fit a full-wind-field fitting line in a rectangular coordinate system with the ambient temperature and the bearing temperature as the X-axis and Y-axis respectively, use it as the initial threshold line corresponding to each wind farm, and determine the slope of each initial threshold line.

[0104] Then, take the average value of the slopes of all initial threshold lines as the slope of the bearing temperature trend threshold line.

[0105] After determining the slope of the bearing temperature trend threshold line by using the above method, the intercept of the preliminary threshold line y = ax in the above rectangular coordinate system can be adjusted by using the second historical monitoring data set to obtain the bearing temperature trend threshold line y = ax + b, so that the sample points determined by the bearing temperature and the ambient temperature collected at each sampling moment in the second historical monitoring data set are evenly distributed on both sides of the bearing temperature trend threshold line y = ax + b.

[0106] Specifically, the bearing temperature collected at each sampling moment in the second historical monitoring data set can be expressed as {x1, x2,... x t}, and the corresponding ambient temperature is expressed as {y1, y2,... y t}, where t is the number of sampling moments in the second historical monitoring data set.

[0107] When the slope a has been determined, an expected intercept b1 of the bearing temperature trend threshold line can be calculated according to x1 and y1. Similarly, b t corresponding to x t and y t is obtained.

[0108] Finally, calculate the average value of b1, b2,... b t as the intercept of the bearing temperature trend threshold line.

[0109] Thus, according to the slope and intercept of the bearing temperature trend threshold line that have been determined, the bearing temperature trend threshold line can be plotted in the above rectangular coordinate system. It should be noted that the method for establishing the bearing temperature trend threshold line provided by the present invention can be migrated and used in other temperature warning models, such as other types of bearing temperature warning models on wind turbines.

[0110] Based on the content of the above embodiments, as an alternative embodiment, in combination with Figure 2 as shown, when the number of historical monitoring data in the second historical monitoring dataset is not greater than the first quantity threshold, it further includes:

[0111] Obtain a third historical monitoring dataset of other wind turbines of the same model as the wind turbine to be measured in the wind farm within the above-mentioned second preset duration under near-full-load operating conditions;

[0112] Take the bearing temperature and ambient temperature collected at each sampling moment in the third historical monitoring dataset as a set of fitting data, and determine the intercept of the bearing temperature trend threshold line.

[0113] The present invention considers a method for constructing a bearing temperature trend threshold line in the case where the wind turbine to be measured may be newly put into use and there is insufficient historical monitoring data.

[0114] First, use the slope calculation method provided in the above embodiments to determine the slope of the bearing temperature trend threshold line.

[0115] When the data volume of the second historical monitoring dataset related to the wind turbine to be measured is insufficient (for example, the input time of the wind turbine to be measured is less than 1 year), the historical operation data of other wind turbines of the same model as the wind turbine to be measured in the same wind farm can be retrieved, and a third historical monitoring dataset within the second preset duration (i.e., 1 day in the above embodiments) under near-full-load operating conditions can be constructed. For example, the operation data within the first 35 days of the past 9 months can be taken to create a third historical monitoring dataset.

[0116] Then, the intercept of the bearing temperature trend threshold line is determined through the bearing temperature and ambient temperature collected at each sampling moment in the third historical monitoring dataset, which will not be elaborated here.

[0117] The method for detecting abnormal bearing temperature of a wind turbine provided by the present invention can warn of abnormal bearing temperature of newly put into use wind turbines with insufficient historical operation data, and can also be achieved through model migration of the same wind farm and the same model, reducing the occurrence of situations where it is impossible to judge due to insufficient data volume.

[0118] Based on the bearing temperature trend threshold line constructed above, the detection of the first bearing temperature state of the wind turbine to be measured can be performed, specifically including but not limited to the following steps:

[0119] Figure 3 It is a schematic diagram of a bearing temperature trend threshold line provided by the present invention. Let the first preset duration be half a year, that is, from August 2021 to January 2022. According to the ambient temperature collected at each sampling moment during this duration in sequence, the corresponding bearing reference temperature is read from the bearing temperature trend threshold line as shown in Figure 3 That is, each ambient temperature collected at a sampling moment corresponds to a bearing reference temperature.

[0120] Figure 4 It is a schematic diagram of the temperature difference between the bearing temperature of the wind turbine and the bearing temperature trend threshold line provided by the present invention. By determining the temperature difference value between the bearing temperature collected at each sampling moment and its corresponding bearing reference temperature, the temperature difference schematic diagram as shown in Figure 4 can be obtained.

[0121] As an optional embodiment, if the time for performing the abnormal detection of the bearing temperature of the wind turbine is January 1, 2022, the current preset sub-duration within the first preset duration can be set as from December 23, 2021 to December 30, 2021; and the historical preset sub-duration within the first preset duration can be set as from November 23, 2021 to December 23, 2021.

[0122] Calculate the temperature difference values related to each sampling point within the current preset sub-duration respectively, and calculate the average value as the first temperature difference average value; and calculate the temperature difference values related to each sampling point within the historical preset sub-duration and calculate the average value as the second temperature difference average value.

[0123] Then, compare the magnitudes of the first temperature difference average value and the second temperature difference average value. If the first temperature difference average value is greater than the second temperature difference average value, it indicates that the temperature rise of the bearing of the wind turbine to be measured within the recent week is positive and greater than a certain threshold a (that is, there is a situation of a large temperature increase), then it can be determined that the first bearing temperature state of the wind turbine to be measured under the current judgment rule is bearing temperature abnormality.

[0124] Based on the content of the above embodiments, as an optional embodiment, in the case where it is determined that the bearing temperature rise of the wind turbine to be measured within the first preset duration is positive, it further includes:

[0125] If the difference between the first temperature difference average value and the second temperature difference average value is greater than the third temperature threshold, determine that the bearing temperature state of the wind turbine to be measured is bearing temperature severe fault;

[0126] If the difference between the first average temperature difference and the second average temperature difference is not greater than the first temperature threshold but greater than the second temperature threshold, it is determined that the first bearing temperature state of the wind turbine to be measured is a slight bearing temperature fault.

[0127] Optionally, if the difference between the first average temperature difference and the second average temperature difference is greater than the first temperature threshold a (e.g., a = 10°C), it can be determined that the first bearing temperature state is a severe bearing temperature fault; if the difference between the first average temperature difference and the second average temperature difference is greater than the second temperature threshold b (e.g., b = 5°C) but less than or equal to the first temperature threshold a, it can be determined that the first bearing temperature state of the wind turbine to be measured is a slight bearing temperature fault.

[0128] It should be noted that the above first temperature threshold a and the second temperature threshold b can be set according to the actual situation.

[0129] The wind turbine bearing temperature anomaly detection method provided by the present invention uses a dynamic bearing temperature trend threshold line, and determines the bearing temperature trend threshold line related to the wind turbine to be measured according to all the data in the wind farm and the historical data of the wind turbine to be measured at the same time, which fully reflects the historical distribution of the ambient temperature and the bearing temperature. In this way, an adaptive bearing temperature trend threshold line is customized for the wind turbine to be measured, with better effect and more capable of reflecting the bearing temperature law of different wind turbines, improving the accuracy and credibility of the final detection result.

[0130] Based on the content of the above embodiments, as an optional embodiment, after determining the first bearing temperature state of the wind turbine to be measured, it further includes:

[0131] Screen out the bearing temperatures collected at each sampling moment of the wind turbine to be measured within the third preset time period from all the monitoring data;

[0132] According to the bearing temperatures collected at each sampling moment, calculate the daily average bearing temperature of the wind turbine to be measured within the third preset time period;

[0133] If at least one of the daily average bearing temperatures is greater than the third temperature threshold and all the daily average bearing temperatures are less than the fourth temperature threshold, it is determined that the second bearing temperature state of the wind turbine to be measured is a slight bearing temperature fault;

[0134] If there is at least one of the daily average bearing temperatures greater than or equal to the fourth temperature threshold, it is determined that the second bearing temperature state of the wind turbine to be measured is a severe bearing temperature fault; the fourth temperature threshold is greater than the third temperature threshold;

[0135] According to the first bearing temperature state and the second bearing temperature state, determine the true bearing temperature state of the wind turbine to be measured.

[0136] Figure 5 This is the second schematic flow diagram of the method for detecting abnormal bearing temperature of a wind turbine provided by the present invention. Taking the part shown in Figure 5 as an example, the present invention provides a method for determining the temperature state of a second bearing, and comprehensively determining the true bearing temperature state of a wind turbine to be tested according to the temperature state of a first bearing and the temperature state of the second bearing.

[0137] Let the third preset duration be n days. Before performing the bearing temperature abnormality detection, it can be pre-judged whether the temperature sensors for detecting the bearing temperature and the ambient temperature are both normal. Among them, the abnormal bearing temperature of the wind turbine caused by the temperature sensor failure should be attributed to the temperature sensor failure. This type of failure does not indicate abnormal bearing temperature and belongs to the premise processing of the model to ensure that the following temperatures involved in the detection are indeed the true temperatures of the bearing and the nacelle.

[0138] When it is determined that the temperature sensors are all normal, first calculate the average bearing temperature for each day within n days (calculate the average value of the bearing temperatures collected at each sampling moment of each day), and then a total of n average bearing temperatures can be obtained.

[0139] If at least one of the average bearing temperatures is greater than the third temperature threshold c (set to 48 °C), but all the average bearing temperatures are less than the fourth temperature threshold d (set to 60 °C), it can be known that the bearing temperature on a certain day (or several days) has an abnormal situation.

[0140] Furthermore, if the average bearing temperature on at least one day within n days is greater than the fourth temperature threshold, it indicates that the bearing has an abnormal bearing temperature situation on this day and is relatively serious.

[0141] In other cases except the above, that is, if the average bearing temperature of each day is less than or equal to the third temperature threshold c, it can be determined that the bearing temperature of the wind turbine to be tested is normal.

[0142] Furthermore, when the temperature states of the first bearing and the second bearing within the second duration are determined, the final state of the bearing temperature state (for the convenience of description, the state here is called the true bearing temperature state in this embodiment) can be determined.

[0143] For example: If one of the results of the temperature states of the first bearing and the second bearing is a serious bearing temperature fault, it is determined that the true bearing temperature state of the wind turbine to be tested is a serious bearing temperature fault; if the results of both are minor bearing temperature faults, it is determined that the final bearing temperature state of the wind turbine to be tested is a minor bearing temperature fault; if the results of both are no faults, it is determined that the true bearing temperature state of the wind turbine to be tested is no fault.

[0144] The present invention provides a method for judging the abnormal state of the bearing temperature of a wind turbine to be measured according to the comparison between the average bearing temperature of the generator set to be measured in each stage of the time sequence and a preset temperature threshold, and the comparison logic is clear.

[0145] Based on the content of the above embodiments, in combination with Figure 5 As shown, as an alternative embodiment, it further includes:

[0146] According to the monitoring data set of all wind turbines in the wind farm under the condition of approaching full load operation, calculate the average bearing temperature and temperature standard deviation of all wind turbines in the wind farm, so as to determine the first allowable range and the second allowable range of bearing temperature according to the average bearing temperature and temperature standard deviation; the second allowable range of bearing temperature is greater than the first allowable range of bearing temperature;

[0147] If the bearing temperature collected at any sampling moment within the fourth preset time period of the wind turbine to be measured exceeds the first allowable range of bearing temperature but is within the second allowable range of bearing temperature, it is determined that the third bearing temperature state is a slight bearing temperature fault;

[0148] If the bearing temperature collected at any sampling moment within the fourth preset time period of the wind turbine to be measured exceeds the second allowable range of bearing temperature, determine the third bearing temperature state of the wind turbine to be measured;

[0149] According to the third bearing temperature state and the true bearing temperature state, determine the final bearing temperature state of the wind turbine to be measured.

[0150] The present invention provides another rule for detecting abnormal bearing temperature of a wind turbine, that is, making a horizontal comparison between the bearing temperature of the wind turbine to be measured within a certain preset time period and the average bearing temperature of other wind turbines in the same wind farm.

[0151] For example: First, calculate the bearing temperature collected at each sampling moment within the fourth preset time period (such as 10 days) of all other wind turbines in the entire wind farm, and then calculate the average value to obtain the average bearing temperature within this time period.

[0152] Then, calculate the average bearing temperature r of all other wind turbines in the wind farm within 10 days, and calculate the relevant temperature standard deviation σ.

[0153] Optionally, let the first allowable range of the bearing temperature be [r - σ, r + σ], and let the second allowable range of the bearing temperature be [r - 3σ, r + 3σ]. If the bearing temperature collected at each sampling moment of the wind turbine to be measured in the recent 10 days is within the first allowable range of the bearing temperature, it is considered that the bearing temperature of the wind turbine to be measured is stable and normal; if among the bearing temperatures collected at each sampling moment of the wind turbine to be measured in the recent 10 days, at least one sampling moment's collected bearing temperature exceeds the first allowable range of the bearing temperature but is within the second allowable range of the bearing temperature, it is considered that there is a minor bearing temperature fault; if among the bearing temperatures collected at each sampling moment of the wind turbine to be measured in the recent 10 days, at least one sampling moment's collected bearing temperature exceeds the second allowable range of the bearing temperature, it is considered that there is a severe bearing temperature fault.

[0154] As another alternative embodiment, it is also possible to calculate the temperature difference value between the bearing temperature at each sampling moment of the wind turbine to be measured within 10 days and the bearing reference temperature determined from the bearing temperature trend threshold line, and finally determine the average temperature difference value within 10 days (which can be called the temperature rise).

[0155] Similarly, obtain the average temperature difference value within 10 days of all other wind turbines in the unified wind farm. Then, by replacing the temperature average value in the step of obtaining the third bearing temperature state result with the above average temperature difference value, a new third bearing temperature state is obtained.

[0156] The method for detecting abnormal bearing temperature of a wind turbine provided by the present invention provides another detection rule by horizontally comparing the bearing temperature of the wind turbine to be measured with the bearing temperatures of other wind turbines in the unified wind farm, and realizes the detection of abnormal bearing temperature from multiple perspectives, which can further improve the accuracy of detection.

[0157] Based on the content of the above embodiment, determining the final bearing temperature state of the wind turbine to be measured according to the third bearing temperature state and the real bearing temperature state includes:

[0158] If any one of the results of the third bearing temperature state and the final bearing temperature state is a severe bearing temperature fault, it is determined that the final bearing temperature state of the wind turbine to be measured is a severe bearing temperature fault;

[0159] If neither of the results of the third bearing temperature state and the final bearing temperature state is a severe bearing temperature fault, but any one of the results is a minor bearing temperature fault, it is determined that the final bearing temperature state of the wind turbine to be measured is a minor bearing temperature fault;

[0160] Otherwise, it is determined that the final bearing temperature state of the wind turbine to be measured is normal.

[0161] In summary, considering the safety of the operation of the wind turbine, the present invention takes the highest fault level among the first bearing temperature state, the second bearing temperature state, and the third bearing temperature state as the final bearing temperature state, and can also combine the abnormal reminders as the detailed description of the model for the unit to be tested.

[0162] If the results of the first bearing temperature state, the second bearing temperature state, and the third bearing temperature state are all normal, the final bearing temperature state of the wind turbine to be tested is set to normal, and the diagnosis process ends.

[0163] Figure 6 is a schematic structural diagram of the wind turbine bearing temperature abnormality detection device provided by the present invention, as Figure 6 shown, mainly including a data acquisition unit 61, a data screening unit 62, a temperature analysis unit 63, and a temperature analysis unit 64, where:

[0164] The data acquisition unit 61 is used to obtain the monitoring data set of all wind turbines in the wind farm where the wind turbine to be tested is located under the near-full-load operation condition, and the monitoring data set includes the bearing temperature and the ambient temperature collected at each sampling moment.

[0165] The data screening unit 62 is used to screen out the bearing temperature and the ambient temperature collected at each sampling moment of the wind turbine to be tested within the first preset time period from all the monitoring data.

[0166] The temperature analysis unit 63 is used to, for the bearing temperature and the ambient temperature collected at any sampling moment, determine the corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculate the temperature difference value between the bearing temperature and the bearing reference temperature.

[0167] The result prediction unit 64 is used to determine the first bearing temperature state of the wind turbine to be tested according to the temperature difference values corresponding to each sampling moment within the first preset time period.

[0168] It should be noted that the wind turbine bearing temperature abnormality detection device provided by the embodiments of the present invention can execute the wind turbine bearing temperature abnormality detection method described in any of the above embodiments during specific operation, and this embodiment will not be elaborated herein.

[0169] The wind turbine bearing temperature abnormality detection device provided by the present invention can accurately judge the bearing temperature rise of the wind turbine according to the temperature difference value between the bearing temperature and the bearing temperature trend threshold line at each current sampling moment, take it as the reference standard for judging the bearing temperature state, can deeply utilize the bearing temperature rise law of the wind turbine, improves the accuracy and credibility of the bearing temperature abnormality detection result, and has a longer early warning time.

[0170] Figure 7 is a schematic structural diagram of the electronic device provided by the present invention. As Figure 7 shown, the electronic device may include: a processor 710, a communications interface 720, a memory 730, and a communication bus 740. Among them, the processor 710, the communications interface 720, and the memory 730 complete mutual communication through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the method for detecting abnormal bearing temperature of a wind turbine, and the method includes:

[0171] Obtain the monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under the condition of approaching full load operation. The monitoring data set includes the bearing temperature and the ambient temperature collected at each sampling moment; screen out the bearing temperature and the ambient temperature collected at each sampling moment of the wind turbine to be measured within the first preset time period from all the monitoring data; for the bearing temperature and the ambient temperature collected at any sampling moment, determine the corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculate the temperature difference value between the bearing temperature and the bearing reference temperature; determine the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset time period.

[0172] In addition, when the logic instructions in the above-mentioned memory 730 are implemented in the form of software function units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0173] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the wind turbine bearing temperature anomaly detection method provided by the above-mentioned various methods. The method includes: obtaining a monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under near-full load operating conditions. The monitoring data set includes the bearing temperature and ambient temperature collected at each sampling moment; screening out the bearing temperature and ambient temperature collected at each sampling moment by the wind turbine to be measured within a first preset time period from all the monitoring data; for the bearing temperature and ambient temperature collected at any sampling moment, determining the corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculating the temperature difference value between the bearing temperature and the bearing reference temperature; determining the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset time period.

[0174] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is configured to execute the wind turbine bearing temperature anomaly detection method provided by the above-mentioned various embodiments. The method includes: obtaining a monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under near-full load operating conditions. The monitoring data set includes the bearing temperature and ambient temperature collected at each sampling moment; screening out the bearing temperature and ambient temperature collected at each sampling moment by the wind turbine to be measured within a first preset time period from all the monitoring data; for the bearing temperature and ambient temperature collected at any sampling moment, determining the corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculating the temperature difference value between the bearing temperature and the bearing reference temperature; determining the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset time period.

[0175] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0176] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting abnormal bearing temperature of a wind turbine, characterized in that, Including: Obtain a monitoring data set of all wind turbines in the wind farm where the wind turbine to be tested is located under near-full-load operating conditions. The monitoring data set includes the bearing temperature and ambient temperature collected at each sampling moment. Screen out the bearing temperature and ambient temperature collected at each sampling moment of the wind turbine to be tested within a first preset time period from all the monitoring data. For the bearing temperature and ambient temperature collected at any sampling moment, determine the corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculate the temperature difference value between the bearing temperature and the bearing reference temperature. The method for constructing the bearing temperature trend threshold line includes: Based on the first historical monitoring data set of all wind turbines in the wind farm under near-full-load operating conditions, determine the slope of the bearing temperature trend threshold line. Obtain a second historical monitoring data set of the wind turbine to be tested under near-full-load operating conditions within a second preset time period recently. Determine the intercept of the bearing temperature trend threshold line according to the second historical monitoring data set of the wind turbine to be tested recently under near-full-load operating conditions. Determine the first bearing temperature state of the wind turbine to be tested according to the temperature difference values corresponding to each sampling moment within the first preset time period.

2. The method for detecting abnormal bearing temperature of a wind turbine according to claim 1, characterized in that, The determining the first bearing temperature state of the wind turbine to be tested according to the temperature difference values corresponding to each sampling moment within the first preset time period includes: Obtain the average value of the temperature difference values at all sampling moments of the wind turbine to be tested within a current preset sub-time period within the first preset time period as the first temperature difference average value. Obtain the average value of the temperature difference values at all sampling moments of the wind turbine to be tested within a historical preset sub-time period within the first preset time period as the second temperature difference average value. If the first temperature difference average value is greater than the second temperature difference average value, it is determined that the bearing temperature rise of the wind turbine to be tested within the first preset time period is positive, and then it is determined that the first bearing temperature state of the wind turbine to be tested is abnormal bearing temperature.

3. The method for detecting abnormal bearing temperature of a wind turbine according to claim 2, wherein In the case of determining that the bearing temperature rise of the wind turbine to be tested within the first preset time period is positive, it further includes: If the difference between the first temperature difference average value and the second temperature difference average value is greater than a third temperature threshold, it is determined that the bearing temperature state of the wind turbine to be tested is a serious bearing temperature fault. If the difference between the first temperature difference average value and the second temperature difference average value is not greater than a first temperature threshold but greater than a second temperature threshold, it is determined that the first bearing temperature state of the wind turbine to be tested is a minor bearing temperature fault.

4. The method for detecting abnormal bearing temperature of a wind turbine according to any one of claims 1-3, characterized in that, The bearing temperature trend threshold line is pre-constructed in the following manner: Obtain the first historical monitoring data set of all wind turbines in the wind farm under near-full-load operating conditions; the first historical monitoring data set includes the bearing temperature and ambient temperature of each wind turbine collected at multiple historical sampling moments. Construct a rectangular coordinate system with the ambient temperature and bearing temperature as the X-axis and Y-axis respectively, and mark the bearing temperature and ambient temperature collected at each historical sampling moment as a sample point in the rectangular coordinate system. Perform a linear fit on all sample points to obtain the slope of the initially obtained threshold line from the fit as the slope of the bearing temperature trend threshold line; When the number of historical monitoring data in the second historical monitoring dataset is greater than the first quantity threshold, use the bearing temperature and ambient temperature collected at each sampling moment in the second historical monitoring dataset as a set of fitting data, and in combination with the slope of the bearing temperature trend threshold line, calculate an expected intercept of the bearing temperature trend threshold line; Use the average value of the expected intercepts calculated for all sampling moments as the intercept of the bearing temperature trend threshold line.

5. The method for detecting abnormal bearing temperature of a wind turbine according to claim 4, wherein, When the number of historical monitoring data in the second historical monitoring dataset is not greater than the first quantity threshold, it further includes: Obtain a third historical monitoring dataset of other wind turbines of the same model as the wind turbine to be measured in the wind farm within the second preset time period under near-full-load operating conditions; Use the bearing temperature and ambient temperature collected at each sampling moment in the third historical monitoring dataset as a set of fitting data to determine the intercept of the bearing temperature trend threshold line.

6. The method for detecting abnormal bearing temperature of a wind turbine according to claim 1, wherein It further includes: Screen out the bearing temperature collected at each sampling moment by the wind turbine to be measured within the third preset time period from all monitoring data; Based on the bearing temperature collected at each sampling moment, calculate the daily average bearing temperature of the wind turbine to be measured within the third preset time period; If at least one of the daily average bearing temperatures is greater than the third temperature threshold and all daily average bearing temperatures are less than the fourth temperature threshold, determine that the second bearing temperature state of the wind turbine to be measured is a minor bearing temperature fault; If there is at least one of the daily average bearing temperatures greater than or equal to the fourth temperature threshold, determine that the second bearing temperature state of the wind turbine to be measured is a serious bearing temperature fault; the fourth temperature threshold is greater than the third temperature threshold; Based on the first bearing temperature state and the second bearing temperature state, determine the true bearing temperature state of the wind turbine to be measured.

7. The method for detecting abnormal bearing temperature of a wind turbine according to claim 6, wherein It further includes: Based on the monitoring datasets of all wind turbines in the wind farm under near-full-load operating conditions, calculate the average bearing temperature and temperature standard deviation of all wind turbines in the wind farm, and based on the average bearing temperature and temperature standard deviation, determine the first allowable range and the second allowable range of bearing temperature; the second allowable range of bearing temperature is greater than the first allowable range of bearing temperature; If the bearing temperature collected at any sampling moment by the wind turbine to be measured within the fourth preset time period exceeds the first allowable range of bearing temperature but is within the second allowable range of bearing temperature, determine that the third bearing temperature state is a minor bearing temperature fault; If the bearing temperature collected at any sampling moment by the wind turbine to be measured within the fourth preset time period exceeds the second allowable range of bearing temperature, determine the third bearing temperature state of the wind turbine to be measured; Based on the third bearing temperature state and the true bearing temperature state, determine the final bearing temperature state of the wind turbine to be measured.

8. The method for detecting abnormal bearing temperature of a wind turbine according to claim 7, characterized in that, Determining the final bearing temperature state of the wind turbine to be measured according to the third bearing temperature state and the true bearing temperature state includes: If any result of the third bearing temperature state and the final bearing temperature state is a serious bearing temperature fault, it is determined that the final bearing temperature state of the wind turbine to be measured is a serious bearing temperature fault; If neither the result of the third bearing temperature state nor the result of the final bearing temperature state is a serious bearing temperature fault, but any result is a minor bearing temperature fault, it is determined that the final bearing temperature state of the wind turbine to be measured is a minor bearing temperature fault; Otherwise, it is determined that the final bearing temperature state of the wind turbine to be measured is normal.

9. An abnormal bearing temperature detection device for a wind turbine, characterized in that, Including: A data acquisition unit for obtaining a monitoring data set of all wind turbines in the wind farm where the wind turbine to be measured is located under near-full load conditions. The monitoring data set includes the bearing temperature and ambient temperature collected at each sampling moment; A data screening unit for screening out the bearing temperature and ambient temperature collected at each sampling moment of the wind turbine to be measured within a first preset time period from all the monitoring data; A temperature analysis unit for, for the bearing temperature and ambient temperature collected at any sampling moment, determining a corresponding bearing reference temperature from the bearing temperature trend threshold line according to the ambient temperature, and calculating the temperature difference value between the bearing temperature and the bearing reference temperature; The construction method of the bearing temperature trend threshold line includes: Based on the first historical monitoring data set of all wind turbines in the wind farm under near-full load conditions, determining the slope of the bearing temperature trend threshold line; Obtaining a second historical monitoring data set of the wind turbine to be measured under near-full load conditions within a second preset time period recently; Determining the intercept of the bearing temperature trend threshold line according to the second historical monitoring data set of the wind turbine to be measured under near-full load conditions recently; A result prediction unit for determining the first bearing temperature state of the wind turbine to be measured according to the temperature difference values corresponding to each sampling moment within the first preset time period.

10. A wind turbine, characterized in that, It includes a wind turbine body, and a detection processor is provided in the wind turbine body; it also includes a memory and a program or instruction stored on the memory and executable on the detection processor. When the program or instruction is executed by the detection processor, the steps of the wind turbine bearing temperature anomaly detection method according to any one of claims 1 to 8 are implemented.

Images