Bearing fault diagnosis method, device and electronic equipment for wind turbine generator set

Abstract

The present application discloses a bearing fault diagnosis method, device, and electronic device for a wind turbine generator set, which can identify bearing vibration monitoring signals with data quality problems, thereby eliminating the situation of false fault alarms caused by the use of bearing vibration monitoring signals with data quality problems. The bearing fault diagnosis method for a wind turbine generator set includes: obtaining a bearing vibration signal sequence collected by a vibration signal acquisition device; identifying whether the bearing vibration signal sequence has data quality abnormalities based on the distribution of the bearing vibration signal sequence in the time domain; and in the case that the bearing vibration signal sequence does not have data quality abnormalities, performing fault diagnosis on the bearing of the wind turbine generator set based on the bearing vibration signal sequence.

Description

Technical Field

[0001] The present application relates to the wind power generation industry, and in particular to a bearing fault diagnosis method, device, and electronic equipment for a wind power generator set. Background Art

[0002] By installing vibration sensors on the bearing components of the wind turbine, the components of the unit can be monitored in real time. By analyzing the data detected by the vibration sensors, the health status of the core components of the wind turbine and the overall health status of the wind turbine can be determined.

[0003] However, the inventors have found that in actual applications, the data detected by the vibration sensor often produces data anomalies due to damage to the vibration sensor, loose sensor connections, abnormal network transmission, electromagnetic interference caused by power supply, etc. The abnormal vibration data will directly affect the analysis results and may lead to misjudgment of faults, such as false alarm of faults or missed alarm of faults. Summary of the invention

[0004] The embodiments of the present application provide a bearing fault diagnosis method, device, and electronic device for a wind turbine generator set, which can identify bearing vibration monitoring signals with data quality problems, thereby eliminating false fault alarms caused by using bearing vibration monitoring signals with data quality problems.

[0005] On the one hand, an embodiment of the present application provides a bearing fault diagnosis method for a wind turbine generator set, the method comprising: acquiring a bearing vibration signal sequence collected by a vibration signal acquisition device; identifying whether there is data quality abnormality in the bearing vibration signal sequence based on the distribution of the bearing vibration signal sequence in the time domain; in the case that there is no data quality abnormality in the bearing vibration signal sequence, performing fault diagnosis on the bearing of the wind turbine generator set based on the bearing vibration signal sequence.

[0006] Optionally, when the data quality anomaly includes data polarization, the bearing vibration signal sequence is identified to have data quality anomaly based on the distribution of the bearing vibration signal sequence in the time domain, including: dividing the signal strength range of the bearing vibration signal sequence into n intervals, where n is an integer greater than 1; among multiple sampling points of the bearing vibration signal sequence, counting the number of sampling points whose signal strength belongs to each interval; when the number of sampling points in the n intervals follows a normal distribution, determining the interval where the expected value of the normal distribution lies to obtain a first interval, and judging whether the bearing vibration signal sequence has data polarization based on the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence.

[0007] Optionally, based on the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence, it is judged whether there is data polarization phenomenon in the bearing vibration signal sequence, including: counting the number of intervals below the first interval among n intervals to obtain a first value; counting the number of intervals above the first interval among n intervals to obtain a second value; based on the first value and the second value, it is judged whether there is data polarization phenomenon in the bearing vibration signal sequence.

[0008] Optionally, based on the first numerical value and the second numerical value, it is determined whether there is a data polarization phenomenon in the bearing vibration signal sequence, including: when the first numerical value is greater than the first multiple of the second numerical value, determining that the bearing vibration signal sequence is an upper side aggregated signal with data polarization phenomenon; when the second numerical value is greater than the second multiple of the first numerical value, determining that the bearing vibration signal sequence is a lower side aggregated signal with data polarization phenomenon.

[0009] Optionally, before dividing the signal intensity range of the bearing vibration signal sequence into n intervals, identifying whether the bearing vibration signal sequence has data quality abnormalities based on the distribution of the bearing vibration signal sequence in the time domain also includes: obtaining the DC component of the bearing vibration signal sequence; when the DC component is zero, determining that there is no data polarization phenomenon in the bearing vibration signal sequence; and when the DC component is not zero, dividing the signal intensity range of the bearing vibration signal sequence into n intervals.

[0010] Optionally, when the data quality anomaly includes an extreme value distribution anomaly, the bearing vibration signal sequence is identified to have data quality anomaly based on the distribution of the bearing vibration signal sequence in the time domain, including: dividing the bearing vibration signal sequence according to the time axis of the bearing vibration signal sequence to obtain m signal sets, where m is an integer greater than 1; determining the extreme value of the signal strength in each signal set respectively; and determining whether the bearing vibration signal sequence has an extreme value distribution anomaly based on the distribution of all extreme values ​​determined from the m signal sets.

[0011] Optionally, based on the distribution of all extreme values ​​determined by the m signal sets, it is determined whether the bearing vibration signal sequence has an extreme value distribution anomaly, including: counting the mode of all extreme values; counting the number of sampling points in the bearing vibration signal sequence whose signal strength is equal to the mode of the extreme values; calculating the proportion of the number of sampling points in the total number of sampling points in the bearing vibration signal sequence; when the proportion exceeds a proportion threshold, it is determined that the bearing vibration signal sequence has an extreme value distribution anomaly.

[0012] On the other hand, an embodiment of the present application provides a bearing fault diagnosis device for a wind turbine generator set, the device comprising: an acquisition module for acquiring a bearing vibration signal sequence collected by a vibration signal acquisition device; an identification module for identifying whether there is data quality abnormality in the bearing vibration signal sequence based on the distribution of the bearing vibration signal sequence in the time domain; and a diagnosis module for performing fault diagnosis on the bearing of the wind turbine generator set based on the bearing vibration signal sequence when there is no data quality abnormality in the bearing vibration signal sequence.

[0013] Optionally, the identification module includes: a segmentation unit, used to segment the signal strength range of the bearing vibration signal sequence into n intervals when the data quality abnormality includes data polarization, wherein n is an integer greater than 1; a first statistical unit, used to count the number of sampling points whose signal strength belongs to each interval among multiple sampling points of the bearing vibration signal sequence; an execution unit, used to determine the interval where the expected value of the normal distribution is located when the number of sampling points in the n intervals obeys a normal distribution, obtain the first interval, and judge whether there is data polarization in the bearing vibration signal sequence according to the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence.

[0014] Optionally, the execution unit includes: a second statistical unit, used to count the number of intervals in the n intervals that are below the first interval to obtain a first value; a third statistical unit, used to count the number of intervals in the n intervals that are above the first interval to obtain a second value; and a judgment unit, used to judge whether there is data polarization phenomenon in the bearing vibration signal sequence based on the first value and the second value.

[0015] Optionally, the judgment unit includes: a first determination unit, used to determine that the bearing vibration signal sequence is an upper side aggregation signal with data polarization phenomenon when the first value is greater than the first multiple of the second value; a second determination unit, used to determine that the bearing vibration signal sequence is a lower side aggregation signal with data polarization phenomenon when the second value is greater than the second multiple of the first value.

[0016] Optionally, the identification module also includes: an acquisition unit, used to acquire the DC component of the bearing vibration signal sequence before dividing the signal strength range of the bearing vibration signal sequence into n intervals; a third determination unit, used to determine that there is no data polarization phenomenon in the bearing vibration signal sequence when the DC component is zero; and the segmentation unit is used to divide the signal strength range of the bearing vibration signal sequence into n intervals when the DC component is not zero.

[0017] Optionally, the identification module includes: a division unit, used to divide the bearing vibration signal sequence according to the time axis of the bearing vibration signal sequence to obtain m signal sets when the data quality abnormality includes an extreme value distribution abnormality, wherein m is an integer greater than 1; a fourth determination unit, used to respectively determine the extreme value of the signal strength in each signal set; and a fifth determination unit, used to determine whether the bearing vibration signal sequence has an extreme value distribution abnormality based on the distribution of all extreme values ​​determined from the m signal sets.

[0018] Optionally, the fifth determination unit includes: a fourth statistical unit, used to count the mode of all extreme values; a fifth statistical unit, used to count the number of sampling points in the bearing vibration signal sequence whose signal strength is equal to the mode of the extreme value; a calculation unit, used to calculate the proportion of the number of sampling points in the total number of sampling points in the bearing vibration signal sequence; and a sixth determination unit, used to determine that there is an extreme value distribution anomaly in the bearing vibration signal sequence when the proportion exceeds a proportion threshold.

[0019] On the other hand, an embodiment of the present application provides an electronic device, which includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the bearing fault diagnosis method of the wind turbine generator set as described in the embodiment of the present application is implemented.

[0020] On the other hand, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the bearing fault diagnosis method for a wind turbine generator set as described in the embodiment of the present application is implemented.

[0021] The bearing fault diagnosis method, device, and electronic device of the wind turbine generator set in the embodiments of the present application identify whether there is data quality abnormality in the bearing vibration signal sequence through the distribution of the bearing vibration signal sequence in the time domain, and perform bearing fault diagnosis on the wind turbine generator set according to the bearing vibration signal sequence when there is no data quality abnormality in the bearing vibration signal sequence. It is capable of identifying bearing vibration monitoring signals with data quality problems, and when there is an abnormality in the bearing vibration monitoring signal, abandoning the use of the bearing vibration monitoring signal with the abnormality for fault diagnosis, thereby eliminating false fault alarms caused by the use of bearing vibration monitoring signals with data quality problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0023] Figure 1It is an optional schematic diagram of the bearing vibration signal sequence under normal conditions;

[0024] Figure 2 It is a schematic diagram when there is an optional data quality anomaly in the bearing vibration signal sequence;

[0025] Figure 3 It is a schematic diagram when another optional data quality anomaly exists in the bearing vibration signal sequence;

[0026] Figure 4 It is a schematic diagram when another optional data quality anomaly exists in the bearing vibration signal sequence;

[0027] Figure 5 It is a flow chart of a bearing fault diagnosis method for a wind turbine generator set provided by an embodiment of the present application;

[0028] Figure 6 is a flow chart of a bearing fault diagnosis method for a wind turbine generator set provided by another embodiment of the present application;

[0029] Figure 7 is a flow chart of a bearing fault diagnosis method for a wind turbine generator set provided by another embodiment of the present application;

[0030] Figure 8 is a flow chart of a bearing fault diagnosis method for a wind turbine generator set provided by another embodiment of the present application;

[0031] Fig. 9 is a structural schematic diagram of a bearing fault diagnosis device for a wind turbine generator set provided by another embodiment of the present application;

[0032] Fig.10 is a structural schematic diagram of an electronic device provided by another embodiment of the present application;

[0033] Fig.11 It is a schematic diagram of an application scenario of a bearing fault diagnosis device for a wind turbine generator set provided by an embodiment of the present application. DETAILED DESCRIPTION

[0034] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0035] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, 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 includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0036] First, an optional application scenario of the bearing fault diagnosis method for a wind turbine generator set provided in an embodiment of the present application is introduced below.

[0037] The bearing fault diagnosis method of a wind turbine provided in the embodiment of the present application can be applied in an application environment for monitoring the bearings of a wind turbine. Specifically, the components of the wind turbine include bearings, and the vibration of the bearings can reflect the operation of the core components of the wind turbine, as well as the overall health of the operation of the wind turbine, etc. Therefore, the vibration of the bearings can be monitored in real time to obtain a vibration monitoring signal of the bearings. A vibration signal acquisition device can be installed in the wind turbine, wherein the vibration signal acquisition device can include a vibration sensor for sensing the vibration intensity, and the vibration sensor can be installed on the bearings.

[0038] The vibration signal acquisition device may further include a signal processing module connected to the vibration sensor, etc. The signal processing module connected to the vibration sensor may be connected to the vibration sensor by wired cable communication or wireless communication. The signal processing module may be used to receive the signal obtained by the vibration sensor sensing the vibration intensity, convert the analog signal into a digital signal, and the signal of each sampling point is used to indicate the magnitude of the vibration intensity collected at the corresponding sampling moment.

[0039] Optionally, the signal processing module connected to the vibration sensor can be installed in the hub, cabin and other parts of the wind turbine generator set, and the present application embodiment does not impose specific restrictions on this. The vibration sensor samples at a certain sampling frequency to obtain the signal of the bearing during the vibration process. When the signal is normal and there is no data quality abnormality, an optional schematic diagram of the bearing vibration signal sequence is as follows Figure 1 As shown. The vertical axis is the signal quantity, which can be a digital quantity collected by the vibration sensor, used to represent the vibration intensity; the horizontal axis is the time axis, and the sampling time of each sampling point is the time of the projection position of the corresponding sampling point on the horizontal axis.

[0040] Due to damage to the vibration sensor, loose sensor connections, abnormal network transmission, electromagnetic interference caused by power supply, etc., the monitored bearing vibration signal sequence may have abnormal data (signal quantity) quality, for example, data polarization (or unilateral aggregation) phenomenon occurs. The types of data polarization include upper side aggregation and lower side aggregation, or flat peak phenomenon occurs. Among them, a unilaterally aggregated signal means that for a signal sequence presenting a sine waveform or a cosine waveform, the waveform on the first side (such as the upper side or the lower side) is normal, while the signal quantity on the other side is concentrated together and does not present a waveform similar to the first side; such a signal is called a unilaterally aggregated signal. A signal in which the signal quantities of the sampling points are gathered together on the lower side is called a lower side aggregated signal. An optional schematic diagram is shown as follows. Figure 2 As shown; a section of the signal where the signal quantities of the sampling points are gathered together on the upper side is called the upper side gathered signal. An optional schematic diagram is shown as Figure 3 As shown. A flat peak signal means that after a signal sequence is segmented on the horizontal axis, there is at least one sampling time segment: in this time segment, there are many sampling points whose signal quantity is equal to the maximum or minimum value of the signal in this time segment, and even some sampling points are continuous and unchanged, and are continuously equal to the maximum or minimum value of the signal. Then, such a signal sequence can be called a flat peak signal. An optional schematic diagram is shown as follows Figure 4 As shown in Figure 2, the flat peak signal has the problem of abnormal distribution of the maximum value.

[0041] The bearing vibration signal collected by the vibration signal collection device can be sent to the monitoring system, and the monitoring system analyzes the health status of each core component of the wind turbine and the overall health status of the wind turbine according to the bearing vibration signal. The specific fault diagnosis / health status analysis method is not repeated here. The monitoring system can be a software program running in an electronic device with computing power, and performs fault diagnosis according to the received vibration signal. The above-mentioned electronic device can be an electronic device provided in an embodiment of the present application, and the electronic device can include: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the bearing fault diagnosis method of the wind turbine described in the embodiment of the present application is implemented. The above-mentioned electronic device can be arranged inside the nacelle, tower and other components of the wind turbine, or it can also be arranged in the monitoring room of the wind farm where the wind turbine is located, etc., and the embodiment of the present application does not make specific limitations on this.

[0042] Optionally, the monitoring system can be a content management system (CMS) commonly used in the wind power industry. CMS mainly monitors the components of the wind turbine in real time by installing vibration sensors on the blades or bearings of the wind turbine, identifies the health status of the core components of the unit and the overall health status of the unit through vibration data analysis, and arranges the operation and maintenance plan reasonably according to the health status of the components. In practice, the bearing vibration signal acquisition data acquired by CMS often causes abnormal vibration data due to damage to the vibration sensor, loose sensor connection, abnormal network transmission, electromagnetic interference caused by power supply, etc. Abnormal vibration data will directly affect the vibration analysis results, causing CMS to falsely report faults or miss faults, which brings bad experience and trouble to the operation and maintenance personnel. It can be seen that real-time detection of CMS data quality abnormalities and timely giving of data quality abnormality prompts can avoid on-site operation and maintenance personnel from misjudging the CMS diagnosis results due to data quality problems, which is of great significance to the practical application of CMS.

[0043] In order to eliminate the interference of abnormal data quality of bearing vibration signals on bearing fault diagnosis, an embodiment of the present application provides a bearing fault diagnosis method for a wind turbine generator set. Optionally, the method provided in the embodiment of the present application can be integrated into the above-mentioned CMS system.

[0044] Figure 5 FIG. 1 is a flow chart of a method for diagnosing bearing faults of a wind turbine generator set provided by an embodiment of the present application. Figure 5 As shown, the method comprises the following steps:

[0045] Step 101, obtaining a bearing vibration signal sequence collected by a vibration signal collection device.

[0046] The vibration signal collection device may include a vibration sensor for collecting the vibration of the bearing in the monitoring direction at a certain signal sampling frequency, and the collected signal quantity is used to represent the monitored vibration intensity.

[0047] The bearing vibration signal sequence includes a sequence of multiple signals collected at a preset sampling frequency within a period of time, which can also be called multiple sampling points. The time interval between each sampling point is a sampling period, and the signal strength of each sampling point is the vibration strength at the corresponding sampling moment.

[0048] Step 102: Identify whether the bearing vibration signal sequence has data quality abnormalities according to the distribution of the bearing vibration signal sequence in the time domain.

[0049] The signal with abnormal data quality may be the above-mentioned upper side aggregated signal, lower side aggregated signal or flat peak signal. Among them, there is data polarization in the upper side aggregated signal and the lower side aggregated signal, and the signal quantity of the sampling point tends to be aggregated on the upper side or the lower side. The flat peak signal has an abnormal maximum value distribution, and the number of signal quantities equal to the maximum value or the minimum value is large (higher than the preset number).

[0050] According to the above definition of data quality anomaly, the distribution of the bearing vibration signal sequence in the time domain can be analyzed to identify whether the bearing vibration signal sequence has data quality anomaly.

[0051] Step 103: When there is no data quality abnormality in the bearing vibration signal sequence, fault diagnosis is performed on the bearing of the wind turbine generator set according to the bearing vibration signal sequence.

[0052] When it is determined according to step 102 that there is no data quality abnormality in the bearing vibration signal sequence, fault diagnosis of the bearing of the wind turbine generator set may be performed according to the bearing vibration signal sequence.

[0053] An optional example is that the CMS system can analyze the operating health of the main core components (including bearings) of the wind turbine generator set based on the bearing vibration signal sequence.

[0054] If any data quality abnormality exists in the bearing vibration signal sequence, the collected part of the bearing vibration signal sequence will be abandoned. Otherwise, if fault diagnosis is performed based on the bearing vibration signal sequence with abnormal data quality, it may lead to erroneous diagnosis of false fault alarms or missed faults.

[0055] In addition, when there is data quality abnormality in the bearing vibration signal sequence, a prompt alarm can be issued to prompt the operation and maintenance personnel to detect the path for collecting vibration signals, including detecting the inside of the vibration signal acquisition device, such as whether the vibration sensor can work normally, whether the communication between the vibration sensor and the signal processing module of the vibration signal acquisition device is normal, and whether the communication between the vibration signal acquisition device and the monitoring system is normal. After the maintenance is normal, the bearing vibration signal sequence is collected again, and the method provided in the embodiment of the present application is executed. If a bearing vibration signal sequence without data quality abnormality is obtained, the bearing of the wind turbine generator set can be further diagnosed for faults.

[0056] For step 102, the existing methods for detecting data quality anomalies in other fields, such as testing the stability and periodicity of data, identifying whether the data is abnormal, etc., cannot be directly applied to the detection of complex vibration data anomalies. This is because the vibration data itself is not stable data, and the periodicity of the data will be different when different faults occur in the equipment. Therefore, there is currently a lack of effective detection methods for vibration data anomalies. In order to solve data quality problems such as data polarization and extreme value distribution anomalies, it is necessary to design an algorithm that is more suitable for application scenarios of monitoring vibration signals.

[0057] In a possible implementation, the above data quality anomaly includes data polarization phenomenon, such as Figure 2 The data shown is the data polarization phenomenon and the Figure 3 The data shown is a data polarization phenomenon in which the data are concentrated on the upper side. In this case, whether there is a data polarization phenomenon can be determined by the position of the interval where the signal strength is concentrated within the total signal strength range.

[0058] An optional implementation manner is that the above step 102 identifies whether the bearing vibration signal sequence has data quality abnormality according to the distribution of the bearing vibration signal sequence in the time domain, specifically comprising executing the following steps:

[0059] Step 201: divide the signal intensity range of the bearing vibration signal sequence into n intervals, where n is an integer greater than 1.

[0060] The signal strength range of the bearing vibration signal sequence is the range [min, max] between the maximum signal strength max and the minimum signal strength min of the bearing vibration signal sequence. When divided into n intervals, it can be divided according to the preset number of intervals, that is, n is a pre-specified value. Alternatively, it can also be divided according to the preset step size step, and correspondingly, n is a value determined according to the signal strength range and the step size: n = (max-min) / step.

[0061] Step 202: among the multiple sampling points of the bearing vibration signal sequence, the number of sampling points whose signal strength belongs to each interval is counted to obtain n numerical values, each numerical value being used to represent the number of sampling points whose signal strength falls within the corresponding interval.

[0062] For example, for Figure 2The bearing vibration signal sequence shown in the figure determines that the signal strength range of the signal sequence on the vertical axis is [a, b]. The signal strength range [a, b] is divided into n intervals, and the length of each interval on the vertical axis is (ba) / n. Optionally, the above is a method for dividing intervals in which the number of intervals n is a preset number. Another way to divide the intervals is to preset a step size. Specifically, a step size step between partitions can be preset, and then [a, b] is divided into multiple intervals according to the step size step. Then, n = (ba) / step. Then, after dividing the intervals, the number of sampling points whose signal strength falls within each interval is counted separately to obtain n sampling point numbers. Each sampling point number is used to represent the number of sampling points whose signal strength belongs to the corresponding interval.

[0063] Step 203, when the number of sampling points in n intervals obeys a normal distribution, determine the interval in which the expected value μ of the normal distribution is located in the n intervals to obtain the first interval, and judge whether there is data polarization in the bearing vibration signal sequence based on the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence.

[0064] If the bearing vibration signal sequence is normal, such as Figure 1 As shown, the signal intensity of the bearing vibration signal sequence should be evenly distributed in each interval, and the number of n sampling points does not obey the normal distribution; if the signal has a unilateral aggregation (data polarization) phenomenon, such as Figure 2 or Figure 3 As shown, the number of sampling points in n intervals follows a normal distribution.

[0065] Normal distribution, also known as Gaussian distribution. Normal distribution is defined as if the random variable X obeys a mathematical expectation of μ and variance of σ. 2 The probability distribution of is, and its probability density function is

[0066]

[0067] Then this random variable is called a normal random variable, and the distribution it obeys is called a normal distribution, denoted by X~N(μ,σ 2 ). The expected value μ determines its position, and its standard deviation σ determines the amplitude of the distribution. When μ=0, σ=1, the normal distribution is the standard normal distribution.

[0068] To determine whether the number of sampling points in n intervals (which can be regarded as a numerical sequence Q containing n elements) obeys a normal distribution, some mathematical analysis software, such as Matlab, etc., can be used. Alternatively, the determination of whether a sequence obeys a normal distribution can be realized through a software program.

[0069] Furthermore, when it is determined that Q does not obey the normal distribution, it can be determined that there is no data polarization phenomenon in the bearing vibration signal sequence; if it is determined that Q obeys the normal distribution, the following steps can be further performed: first, among the n intervals, determine the first interval where the μ value of the normal distribution is located; secondly, based on the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence, determine whether there is data polarization phenomenon in the bearing vibration signal sequence.

[0070] Specifically, the number of intervals below the first interval among the n intervals can be counted to obtain the first value P1, and the number of intervals above the first interval among the n intervals can be counted to obtain the second value. Then, based on P1 and P2, it can be judged whether there is data polarization in the bearing vibration signal sequence.

[0071] Among them, an optional judgment condition for the data polarization phenomenon is that when the first value is greater than the first multiple of the second value, the bearing vibration signal sequence is determined to be an upper side aggregated signal with data polarization phenomenon; when the second value is greater than the second multiple of the first value, the bearing vibration signal sequence is determined to be a lower side aggregated signal with data polarization phenomenon.

[0072] Optionally, before executing step 201 to divide the signal strength range of the bearing vibration signal sequence into n intervals, step 102 may further include the following steps to identify whether the bearing vibration signal sequence has data quality abnormalities according to the distribution of the bearing vibration signal sequence in the time domain:

[0073] Step 204, obtaining a DC component of the bearing vibration signal sequence;

[0074] Step 205, when the DC component is zero, determining that the bearing vibration signal sequence does not have data polarization phenomenon;

[0075] Step 206: When the DC component is not zero, divide the signal intensity range of the bearing vibration signal sequence into n intervals.

[0076] In signal processing, the value of the DC component can be regarded as an average value, that is, a mathematical expectation, in a statistical sense. Then, whether there is a data polarization phenomenon can be determined based on the DC component. If the DC component is 0, it means that there is no data polarization phenomenon. If the DC component is not 0, there may be a data polarization phenomenon. Steps 201 to 203 can be further performed to determine whether there is a data polarization phenomenon.

[0077] Optionally, in addition to data polarization phenomena, data quality anomalies may also include extreme value distribution anomalies, such as Figure 4In the case where the data quality abnormality includes the abnormal distribution of the extreme values, step 102 identifies whether the bearing vibration signal sequence has data quality abnormality according to the distribution of the bearing vibration signal sequence in the time domain, which may include the following steps:

[0078] Step 301, dividing the bearing vibration signal sequence according to the time axis of the bearing vibration signal sequence to obtain m signal sets, where m is an integer greater than 1;

[0079] Step 302, respectively determining the maximum value of the signal strength in each signal set, including the maximum value and the minimum value;

[0080] Step 303: Determine whether the bearing vibration signal sequence has an abnormal distribution of the maximum values ​​according to the distribution of all the maximum values ​​determined by the m signal sets.

[0081] For step 303, based on the distribution of all the extreme values ​​determined by the m signal sets, it is determined whether the bearing vibration signal sequence has an abnormal distribution of the extreme values. An optional implementation method is to perform the following steps:

[0082] Step 401, counting the mode of all extreme values.

[0083] The extreme value may include the maximum value of the signal strength and / or the minimum value of the signal strength. The mode refers to the value that appears most frequently in a set of data. Sometimes there are several modes in a set of data. The mode can represent the general level of the data and is not affected by extreme data.

[0084] Step 402: Count the number of sampling points in the bearing vibration signal sequence whose signal strength is equal to the maximum value.

[0085] Step 403, calculating the proportion of the number of sampling points in the total number of sampling points in the bearing vibration signal sequence.

[0086] Step 404: when the ratio exceeds the ratio threshold, it is determined that the bearing vibration signal sequence has an abnormal maximum distribution.

[0087] After the bearing vibration signal sequence is divided into m signal sets based on the time axis, the maximum value and / or minimum value of the signal strength in each signal set can be counted to obtain multiple maximum values ​​and / or multiple minimum values. By counting the mode of multiple maximum values ​​and / or multiple minimum values, the most frequently occurring value can be determined.

[0088] Next, in the bearing vibration signal sequence, the sampling points with the same mode as the most frequently occurring maximum values ​​are counted, so that whether the maximum value distribution is abnormal can be determined based on the number of these sampling points. If the ratio of the number of sampling points with the same signal strength as the mode of one of the signal strength maximum values ​​to the total number of sampling points exceeds a preset ratio threshold, the maximum value distribution is determined to be abnormal.

[0089] The bearing fault diagnosis method for a wind turbine generator set in an embodiment of the present application identifies whether there is data quality abnormality in the bearing vibration signal sequence through the distribution of the bearing vibration signal sequence in the time domain, and performs bearing fault diagnosis on the wind turbine generator set according to the bearing vibration signal sequence when there is no data quality abnormality in the bearing vibration signal sequence. It is capable of identifying bearing vibration monitoring signals with data quality problems, and when there is an abnormality in the bearing vibration monitoring signal, abandoning the use of the bearing vibration monitoring signal with the abnormality for fault diagnosis, thereby eliminating false fault alarms caused by the use of bearing vibration monitoring signals with data quality problems.

[0090] Combine the following Figure 6 An optional specific implementation of the bearing fault diagnosis method for a wind turbine generator set provided in an embodiment of the present application is described in detail as follows:

[0091] In this example, the vibration signal acquisition device may include a vibration sensor, a signal processing module and a demodulator. The vibration sensor is installed on the bearing. The collected vibration signal data is converted from analog signal to digital signal through the signal processing module to obtain a bearing vibration signal sequence.

[0092] like Figure 6 As shown, first, the monitoring system acquires the bearing vibration signal sequence collected by the vibration signal collection device.

[0093] In order to avoid misjudgment of faults in the monitoring system due to quality problems of the collected signal data, after obtaining the bearing vibration signal sequence and before performing fault detection based on the bearing vibration signal sequence, the data quality of the bearing vibration signal sequence is tested to determine whether there is any data quality abnormality.

[0094] like Figure 6 As shown, data anomaly detection can be performed by respectively judging by the polarization detection module and the anomaly detection module. The polarization detection module is used to judge whether there is data polarization phenomenon in the bearing vibration signal sequence, and the anomaly detection module is used to judge whether there is an extreme value distribution anomaly in the bearing vibration signal sequence.

[0095] Furthermore, if Figure 6As shown, according to the judgment results of the polarization detection module and the abnormality detection module, it is judged whether the bearing vibration signal sequence is normal. Among them, if the bearing vibration signal sequence does not have the above-mentioned data polarization phenomenon and the extreme value distribution abnormality, the signal data is normal, and the bearing vibration signal sequence can be further converted into an engineering quantity, and fault detection is performed according to the engineering quantity. Before the bearing vibration signal sequence is detected whether the signal data is abnormal, the bearing vibration signal sequence is not subjected to signal preprocessing methods such as denoising, so as to avoid causing the bearing vibration signal sequence with abnormal data quality to be preprocessed into a normal signal sequence, resulting in inaccurate judgment results. If the bearing vibration signal sequence has the above-mentioned data polarization phenomenon or the extreme value distribution abnormality, the bearing vibration signal sequence has data quality abnormality, and the collected bearing vibration signal needs to be abandoned, and fault detection is no longer performed based on the collected bearing vibration signal sequence, and it can be prompted that the collected bearing vibration signal is abnormal, so that the staff can find it in time for maintenance.

[0096] For the above-mentioned polarization detection module, that is, the judgment process of whether the bearing vibration signal sequence has the above-mentioned data polarization phenomenon, an optional flow chart is as follows: Figure 7 As shown, the following steps are included:

[0097] Step 1: Calculate the DC component of the signal x(t) and determine whether it is zero.

[0098] In signal processing, the nth harmonic component F(n) of a continuous periodic signal x(t) is as shown in Equation 2, where T is the period.

[0099]

[0100] The average value μ of the continuous signal x(t) in time T' is defined as:

[0101]

[0102] Generally, in signal processing, the DC component is defined as the zeroth harmonic component of Fourier decomposition, that is, the value of F(n) when n=0. Therefore, within one cycle, the above two equations are completely equal. The so-called DC component is the average value (or mathematical expectation) in statistics.

[0103] Determine whether the DC component μ is 0. If μ=0, it is a normal signal and the next step of determination is not performed. Otherwise, the next step of operation is performed.

[0104] Step 2: Segment the signal strength range of signal x(t) to obtain the random variable X.

[0105] The acquisition of random variables X mainly includes the following processes:

[0106] First, calculate the maximum value max, minimum value min, and variance σ of the signal x(t) 2 As shown in Formula 4, N represents the total number of signals:

[0107]

[0108] Secondly, divide the signal x(t) into bins (segments) in the interval [min, max] according to the preset step size. The typical value of the preset step size step can be 10, and the n bin segments are denoted as w i ,i=0,1,2……,n;

[0109] Finally, count each bin w i The number of signals in a bin, the set of the number of signals in all bins is called the random variable X.

[0110] Step 3: Determine whether the signal x(t) has data polarization.

[0111] First, determine whether the random variable X obeys the mathematical expectation μ and variance σ 2 If it meets the probability distribution of , and its density function is as shown in Formula 1, then the signal x(t) is abnormal.

[0112] Secondly, after determining that the signal x(t) is abnormal, further determine whether the type of the signal abnormality is upper side aggregation or lower side aggregation according to the following steps:

[0113] Step 1, identify the bin w_mode where the average value μ is located;

[0114] Step 2, count the number of bins w_left and w_right below and above the w_mode bin;

[0115] Step 3, determine whether w_right>w_left*co; if so, the signal is a lower side aggregated signal, otherwise the signal x(t) is a normal signal, wherein the typical value of the coefficient co is 2;

[0116] Step 4, determine whether w_left>w_right*co; if so, the signal segment is an upper-side concentrated signal, otherwise the signal x(t) is a normal signal.

[0117] The steps performed by the polarization detection module are an optional example. Optionally, after counting the number of samples in each segment, other judgment algorithms can be used to detect whether the signal is abnormal. For example, it can be calculated whether the ratio of the maximum value to the minimum value of the number of sampling points in all segments is greater than a preset threshold. If so, data polarization exists, otherwise, data polarization does not exist, and so on. The embodiments of the present application do not impose specific restrictions on this.

[0118] The polarization detection module can simplify the judgment process by segmenting the signal strength, and has a certain fault tolerance, thereby improving the accuracy of the judgment.

[0119] For the above-mentioned abnormality detection module, that is, the judgment process of whether the bearing vibration signal sequence has the above-mentioned maximum value distribution abnormality, an optional flowchart is as follows: Figure 8 As shown, the following steps are included:

[0120] Step 1: Segment the signal x(t) along the time axis, take the maximum value in each signal set, and obtain the set X.

[0121] Here, let the sampling frequency of the signal be fs, divide the signal x(t) into segments according to the time length n*(1 / fs), and the typical value of n can be taken as 20. Take the maximum value in each segment, see formula 5.

[0122] max i =max(x(t)) (Formula 5)

[0123] All max i Form a set X, where i = 1, 2, 3...

[0124] Step 2: Find the mode of set X.

[0125] The mode may include a statistical mode, or may include a mode M of the set X calculated according to the Pearson empirical method (see Formula 6 for details).

[0126] M = mean(X)-3*(mean(X)-md) (Formula 6)

[0127] Where md represents the median of set X. mean(X) represents the mean of set X. The mode calculated by Pearson's empirical method is close to the theoretical mode and is often called Pearson's approximate mode. There can be multiple modes obtained by statistics, which can be denoted as M j ,j=1,2,3……

[0128] Step 3: Count the number of signals x(t) equal to the mode, and determine whether there is an extreme value distribution anomaly.

[0129] According to various statistical results of the signal x(t), the signal x(t) is judged. The specific operation steps include:

[0130] Step 1: Count the signal x(t) equal to its mode M j ,j=1,2,3……N is the number of elements j ,j=1,2,3……;

[0131] Step 2: Calculate the number of sampling points N where the signal strength is equal to the modej The ratio of the total number of sampling points N of the signal x(t), max q =max(N j / N) is the same as the threshold a, where a is typically 0.1.

[0132] Step 3, if max q >a, then the signal x(t) is a flat peak signal, otherwise it is a normal signal.

[0133] For the anomaly detection module, the number of samples equal to the maximum or minimum value can also be counted in each signal set, and the number of signal sets whose number of sampling points equal to the maximum or minimum value exceeds the threshold value can be calculated to determine whether there is an extreme value distribution anomaly. If the number of the above signal sets is large and exceeds the preset value, then there is an anomaly, otherwise there is no anomaly. The above is only for exemplary description and does not constitute a limitation on the embodiments of the present application. The anomaly detection module simplifies the process of judging the flat peak signal by dividing the signal into multiple signal sets on the time axis and counting the mode of the maximum and minimum values.

[0134] In the above Figure 6 to Figure 8 In the example shown, after determining that there is no data quality anomaly in the signal sequence, the digital signal can be converted into an engineering quantity and fault detection can be performed, which can effectively avoid misjudgment of the fault detection model due to data quality issues.

[0135] The bearing fault diagnosis method for a wind turbine generator set provided in the embodiment of the present application can be executed by a vibration signal acquisition device or a processor of a monitoring system, is time-effective, and does not require the addition of hardware equipment or expansion of computing resources.

[0136] The embodiment of the present application also provides a bearing fault diagnosis device for a wind generator set, which can be used to execute the bearing fault diagnosis method for a wind generator set provided in the embodiment of the present application. Among them, for the parts not described in detail in the bearing fault diagnosis device for a wind generator set provided in the embodiment of the present application, reference can be made to the relevant description in the bearing fault diagnosis of a wind generator set provided in the embodiment of the present application, and no further description will be given here.

[0137] like Fig. 9 As shown, the bearing fault diagnosis device for a wind turbine generator set provided in an embodiment of the present application includes an acquisition module 11 , an identification module 12 and a diagnosis module 13 .

[0138] Among them, the acquisition module 11 is used to acquire the bearing vibration signal sequence collected by the vibration signal acquisition device; the identification module 12 is used to identify whether there is data quality abnormality in the bearing vibration signal sequence according to the distribution of the bearing vibration signal sequence in the time domain; the diagnosis module 13 is used to perform fault diagnosis on the bearing of the wind turbine generator set according to the bearing vibration signal sequence when there is no data quality abnormality in the bearing vibration signal sequence.

[0139] Fig.11 FIG. 1 is a schematic diagram of an application scenario of a bearing fault diagnosis device for a wind turbine generator set provided by an embodiment of the present application. Fig.11 As shown, the bearing fault diagnosis device 731 of the wind turbine generator set provided in the embodiment of the present application can be integrated into the wind turbine master controller 73, and the wind turbine master controller 73 can be arranged in the nacelle 712 and / or the tower 713 of the wind turbine generator set 71. The embodiment of the present application does not limit the specific installation position of the wind turbine master controller 73. The acquisition module 11 of the bearing fault diagnosis device 731 of the wind turbine generator set can acquire the vibration signal collected by the vibration signal acquisition device 72.

[0140] The vibration signal acquisition device 72 may include a vibration sensor 721 and a signal processing module 722. The vibration sensor 721 may be disposed on the bearing 711 of the wind turbine generator set 71 to collect the vibration signal of the bearing 711. The signal processing module 722 may process the analog vibration signal sensed by the vibration sensor 721 to generate a digital bearing vibration signal sequence.

[0141] Optionally, the bearing fault diagnosis device 731 of the wind turbine generator set can communicate with the server 74 set in the central control room of the wind farm, each wind turbine generator set in the wind farm can be installed with a corresponding wind turbine master controller, and the central control room server 74 can communicate with all wind turbine master controllers in the wind farm. In the embodiment of the present application, the central control room server 74 can be based on the bearing fault diagnosis result of the bearing fault diagnosis device 731 of the wind turbine generator set integrated in the wind turbine master controller 73.

[0142] The bearing fault diagnosis device 731 of the wind turbine generator set can send the judgment result of the identification module 12 to the central control room server 74 to display an alarm prompt on the front-end interface of the central control room server 74, prompting the staff that the vibration signal collected by the vibration signal collection device 72 has abnormal data quality, and reminding the staff to immediately inspect the working status of the vibration signal collection device 72.

[0143] The bearing fault diagnosis device 731 of the wind turbine generator set can also send the bearing fault diagnosis result of the diagnosis module 13 to the central control room server 74, so as to display an alarm prompt on the front-end interface of the central control room server 74, prompting the staff to determine that the bearing is faulty based on the bearing vibration signal sequence, and reminding the staff to inspect the bearing immediately.

[0144] Optionally, the identification module 12 includes: a segmentation unit, used to segment the signal strength range of the bearing vibration signal sequence into n intervals when the data quality abnormality includes data polarization, wherein n is an integer greater than 1; a first statistical unit, used to count the number of sampling points whose signal strength belongs to each interval among multiple sampling points of the bearing vibration signal sequence; an execution unit, used to determine the interval where the expected value of the normal distribution is located when the number of sampling points in the n intervals obeys a normal distribution, obtain the first interval, and judge whether there is data polarization in the bearing vibration signal sequence according to the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence.

[0145] Optionally, the execution unit includes: a second statistical unit, used to count the number of intervals in the n intervals that are below the first interval to obtain a first value; a third statistical unit, used to count the number of intervals in the n intervals that are above the first interval to obtain a second value; and a judgment unit, used to judge whether there is data polarization phenomenon in the bearing vibration signal sequence based on the first value and the second value.

[0146] Optionally, the judgment unit includes: a first determination unit, used to determine that the bearing vibration signal sequence is an upper side aggregation signal with data polarization phenomenon when the first value is greater than the first multiple of the second value; a second determination unit, used to determine that the bearing vibration signal sequence is a lower side aggregation signal with data polarization phenomenon when the second value is greater than the second multiple of the first value.

[0147] Optionally, the identification module 12 also includes: an acquisition unit, used to acquire the DC component of the bearing vibration signal sequence before dividing the signal strength range of the bearing vibration signal sequence into n intervals; a third determination unit, used to determine that there is no data polarization phenomenon in the bearing vibration signal sequence when the DC component is zero; and the segmentation unit is used to divide the signal strength range of the bearing vibration signal sequence into n intervals when the DC component is not zero.

[0148] Optionally, the identification module 12 includes: a division unit, used to divide the bearing vibration signal sequence according to the time axis of the bearing vibration signal sequence to obtain m signal sets when the data quality abnormality includes an extreme value distribution abnormality, wherein m is an integer greater than 1; a fourth determination unit, used to respectively determine the extreme value of the signal strength in each signal set; and a fifth determination unit, used to determine whether the bearing vibration signal sequence has an extreme value distribution abnormality based on the distribution of all extreme values ​​determined from the m signal sets.

[0149] Optionally, the fifth determination unit includes: a fourth statistical unit, used to count the mode of all extreme values; a fifth statistical unit, used to count the number of sampling points in the bearing vibration signal sequence whose signal strength is equal to the mode of the extreme value; a calculation unit, used to calculate the proportion of the number of sampling points in the total number of sampling points in the bearing vibration signal sequence; and a sixth determination unit, used to determine that there is an extreme value distribution anomaly in the bearing vibration signal sequence when the proportion exceeds a proportion threshold.

[0150] The bearing fault diagnosis device of the wind turbine generator set in the embodiment of the present application identifies whether there is any data quality abnormality in the bearing vibration signal sequence through the distribution of the bearing vibration signal sequence in the time domain, and performs fault diagnosis on the bearing of the wind turbine generator set according to the bearing vibration signal sequence when there is no data quality abnormality in the bearing vibration signal sequence. It can identify the bearing vibration monitoring signal with data quality problems, and when there is an abnormality in the bearing vibration monitoring signal, abandon the use of the bearing vibration monitoring signal with the abnormality for fault diagnosis, thereby eliminating the situation of false fault alarm caused by the use of the bearing vibration monitoring signal with data quality problems.

[0151] Fig.10 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application is shown.

[0152] The electronic device may include a processor 501 and a memory 502 storing computer program instructions.

[0153] Specifically, the processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0154] The memory 502 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 502 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 502 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 502 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 502 is a non-volatile solid-state memory.

[0155] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of the present application.

[0156] The processor 501 reads and executes the computer program instructions stored in the memory 502 to implement any one of the bearing fault diagnosis methods for the wind turbine generator set in the above embodiments.

[0157] In one example, the electronic device may further include a communication interface 503 and a bus 510. Figure 3 As shown, the processor 501, the memory 502, and the communication interface 503 are connected via a bus 510 and communicate with each other.

[0158] The communication interface 503 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0159] Bus 510 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 510 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0160] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0161] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal sequence carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0162] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0163] The above reference is according to the method of the embodiment of the present application, the flow chart of the device (system) and the computer program product and / or the block diagram described various aspects of the present application.It should be understood that each square box in the flow chart and / or the block diagram and the combination of each square box in the flow chart and / or the block diagram can be realized by computer program instructions.These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the realization of the function / action specified in one or more square boxes of the flow chart and / or the block diagram.Such a processor can be but is not limited to a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit.It can also be understood that each square box in the block diagram and / or the flow chart and the combination of the square boxes in the block diagram and / or the flow chart can also be realized by the dedicated hardware that performs the specified function or action, or can be realized by the combination of dedicated hardware and computer instructions.

[0164] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A bearing fault diagnosis method for a wind turbine generator set, It is characterized in that include: Acquire a bearing vibration signal sequence collected by a vibration signal collection device; According to the distribution of the bearing vibration signal sequence in the time domain, identifying whether the bearing vibration signal sequence has data quality abnormality; In the case that the bearing vibration signal sequence does not have the data quality abnormality, performing fault diagnosis on the bearing of the wind turbine generator set according to the bearing vibration signal sequence; In the case where the data quality abnormality includes data polarization phenomenon, identifying whether the bearing vibration signal sequence has data quality abnormality according to the distribution of the bearing vibration signal sequence in the time domain includes: Dividing the signal intensity range of the bearing vibration signal sequence into n intervals, where n is an integer greater than 1; Among the multiple sampling points of the bearing vibration signal sequence, counting the number of sampling points whose signal strength belongs to each interval; When the number of sampling points in the n intervals obeys a normal distribution, the interval in which the expected value of the normal distribution is located is determined among the n intervals to obtain a first interval, and based on the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence, it is determined whether the bearing vibration signal sequence has the data polarization phenomenon.

2. The bearing fault diagnosis method of a wind turbine generator set according to claim 1, It is characterized in that The determining, based on the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence, whether the bearing vibration signal sequence has the data polarization phenomenon includes: Counting the number of intervals below the first interval among the n intervals to obtain a first value; Counting the number of intervals above the first interval among the n intervals to obtain a second value; It is determined whether the bearing vibration signal sequence has the data polarization phenomenon according to the first value and the second value.

3. The bearing fault diagnosis method of a wind turbine generator set according to claim 2, It is characterized in that The determining, according to the first value and the second value, whether the bearing vibration signal sequence has the data polarization phenomenon includes: In a case where the first value is greater than a first multiple of the second value, determining that the bearing vibration signal sequence is an upper side aggregated signal having the data polarization phenomenon; When the second value is greater than a second multiple of the first value, the bearing vibration signal sequence is determined to be a lower side aggregation signal with the data polarization phenomenon.

4. The method for diagnosing bearing faults of a wind turbine generator set according to claim 1, It is characterized in that Before dividing the signal strength range of the bearing vibration signal sequence into n intervals, identifying whether the bearing vibration signal sequence has data quality abnormality according to the distribution of the bearing vibration signal sequence in the time domain further includes: Acquiring a DC component of the bearing vibration signal sequence; When the DC component is zero, determining that the bearing vibration signal sequence does not have the data polarization phenomenon; When the DC component is not zero, the signal intensity range of the bearing vibration signal sequence is divided into the n intervals.

5. The method for diagnosing bearing faults of a wind turbine generator set according to claim 1, It is characterized in that In the case where the data quality abnormality includes an extreme value distribution abnormality, identifying whether the bearing vibration signal sequence has a data quality abnormality according to the distribution of the bearing vibration signal sequence in the time domain includes: Dividing the bearing vibration signal sequence according to the time axis of the bearing vibration signal sequence to obtain m signal sets, where m is an integer greater than 1; Determine the maximum value of the signal strength in each signal set respectively; According to the distribution of all the extreme values ​​determined by the m signal sets, it is determined whether the bearing vibration signal sequence has the extreme value distribution anomaly.

6. The method for diagnosing bearing faults of a wind turbine generator set according to claim 5, It is characterized in that Determining whether the bearing vibration signal sequence has the extreme value distribution abnormality according to the distribution of all extreme values ​​determined by the m signal sets includes: Count the mode of all maximum values; Counting the number of sampling points in the bearing vibration signal sequence whose signal strength is equal to the mode of the maximum value; Calculate the proportion of the number of sampling points in the total number of sampling points in the bearing vibration signal sequence; When the ratio exceeds the ratio threshold, it is determined that the bearing vibration signal sequence has the maximum distribution anomaly.

7. A bearing fault diagnosis device for a wind turbine generator set, It is characterized in that The device comprises: An acquisition module, used to acquire a bearing vibration signal sequence acquired by a vibration signal acquisition device; An identification module, used to identify whether the bearing vibration signal sequence has data quality abnormality according to the distribution of the bearing vibration signal sequence in the time domain; A diagnosis module, configured to perform fault diagnosis on the bearing of the wind turbine generator set according to the bearing vibration signal sequence when the bearing vibration signal sequence does not have the data quality abnormality; The identification module comprises: a segmentation unit, configured to segment the signal intensity range of the bearing vibration signal sequence into n intervals when the data quality abnormality includes data polarization, wherein n is an integer greater than 1; A first statistical unit is used to count the number of sampling points whose signal strength belongs to each interval among the multiple sampling points of the bearing vibration signal sequence; An execution unit is used to determine, in the case where the number of sampling points in the n intervals obeys a normal distribution, an interval in which an expected value of the normal distribution is located, to obtain a first interval, and to determine whether the bearing vibration signal sequence has the data polarization phenomenon according to the position of the signal strength range of the first interval in the signal strength range of the bearing vibration signal sequence.

8. An electronic device, It is characterized in that The electronic device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the bearing fault diagnosis method for a wind turbine generator set according to any one of claims 1 to 6 is implemented.

9. A computer storage medium, It is characterized in that The computer storage medium stores computer program instructions, and when the computer program instructions are executed by the processor, the bearing fault diagnosis method for a wind turbine generator set according to any one of claims 1 to 6 is implemented.

Images