CN116221038A

CN116221038A - Bearing operation monitoring method and device based on wind power bearing retainer

Info

Publication number: CN116221038A
Application number: CN202310518604.4A
Authority: CN
Inventors: 郑广会; 赵培振; 郑金宇; 郑金泽
Original assignee: Shandong Golden Empire Precision Machinery Technology Co Ltd
Current assignee: Shandong Golden Empire Precision Machinery Technology Co Ltd
Priority date: 2023-05-10
Filing date: 2023-05-10
Publication date: 2023-06-06
Anticipated expiration: 2043-05-10
Also published as: CN116221038B

Abstract

The application provides a bearing operation monitoring method and device based on a wind power bearing retainer, and belongs to the technical field of wind turbine generator operation supervision. The method first acquires a positioning acquisition data set from a positioning detection device. And analyzing the positioning acquisition data set according to the data acquisition time sequence by a difference data positioning algorithm to determine an operation difference data set of the bearing retainer. And determining the association degree value of each operation difference data set and the vibration signal based on the vibration signal and the abnormal trigger association model of the bearing annular shell from the positioning detection device. And respectively carrying out data correction on each operation difference data set based on each association degree value and the vibration intensity value of the vibration signal so as to obtain a corrected operation data set. And determining the corresponding bearing health state of the bearing retainer according to each operation data set, the vibration signal and the corresponding monitoring and early warning standard so as to send the bearing health state to the user terminal and monitor and early warn the operation of the fan bearing.

Description

Bearing operation monitoring method and device based on wind power bearing retainer

Technical Field

The application relates to the technical field of wind turbine generator operation supervision, in particular to a bearing operation monitoring method and device based on a wind power bearing retainer.

Background

Bearing cage means a bearing part which partly encloses all or part of the rolling elements and moves therewith, to isolate the rolling elements, and usually also to guide them and to retain them in the bearing. The wind driven generator is internally provided with a bearing, and the bearing is internally provided with a large-sized bearing retainer, so that the bearing retainer has very important effects on safe, stable and efficient operation of the unit.

The conventional bearing retainer is often only used for isolating the rolling bodies and maintaining the movement positions of the rolling bodies, and if the data such as the running state, the health condition and the like of the bearing retainer are to be obtained, the data can only be obtained manually. On one hand, labor cost is wasted, on the other hand, data information acquisition is delayed, and real-time supervision on the fan bearing and the retainer is not realized.

Disclosure of Invention

In order to solve the problems, the embodiment of the application provides a bearing operation monitoring method and device based on a wind power bearing retainer.

In one aspect, an embodiment of the application provides a bearing operation monitoring method based on a wind power bearing retainer, and the method is applied to a wind turbine generator; the bearing retainer of the wind turbine generator is provided with at least one auxiliary positioning sensor for positioning the bearing retainer in real time and sending out corresponding positioning signals; the bearing annular shell is provided with a positioning detection device for collecting the positioning signals; the method comprises the following steps:

Acquiring a positioning acquisition data set from the positioning detection device; wherein the positioning acquisition data set at least comprises position data of the auxiliary positioning sensor;

analyzing the positioning acquisition data set according to a data acquisition time sequence by a preset difference data positioning algorithm to determine an operation difference data set of the bearing retainer; the position data in the same operation difference data set are the same as the difference value of the preset standard position data;

determining a correlation degree value of each operation difference data set and the vibration signal based on the vibration signal of the bearing annular shell from the positioning detection device and a preset abnormal trigger correlation model;

based on the association degree value and the vibration intensity value of the vibration signal, respectively carrying out data correction on each operation difference data set to obtain corrected operation data sets; one operation data set at least comprises one operation difference data set, and the corresponding association degree values of the operation difference data sets in the same operation data set are the same;

and determining the corresponding bearing health state of the bearing retainer according to each operation data set, the vibration signal and the corresponding monitoring and early warning standard so as to send the bearing health state to a user terminal and monitor and early warn the operation of the fan bearing.

In one implementation manner of the present application, through a preset differential data positioning algorithm, according to a data acquisition time sequence, the positioning acquisition data set is analyzed to determine an operation differential data set of the bearing retainer, which specifically includes:

dividing the data of the positioning acquisition data set in the same monitoring period into a plurality of operation monitoring analysis sequences according to the monitoring period divided in advance; the monitoring time period is obtained by dividing the monitoring time period according to a preset environmental condition change time period; each operation monitoring analysis sequence comprises a plurality of positioning acquisition data; the positioning acquisition data at least comprise a binary group consisting of time and position data;

dividing each positioning acquisition data in each operation monitoring analysis sequence into a plurality of operation monitoring analysis subsequences through the difference data positioning algorithm, and sequentially determining the median of each operation monitoring analysis subsequence according to a preset rule; the median is obtained by comparing the position data;

generating a median sequence corresponding to each median through an bubbling sequencing algorithm so as to determine a selected median in the median sequence;

determining a difference data comparison binary group of element positions corresponding to the difference data screening values in each operation monitoring analysis sequence based on the selected median, the operation monitoring analysis sequence and the preset difference data screening values;

And comparing the position data in each difference data comparison binary group with a preset grouping threshold interval to generate the operation difference data group according to the comparison result and the difference data comparison binary group.

In one implementation manner of the present application, the comparing the position data in each of the difference data comparison tuples with a preset grouping threshold interval to generate the operation difference data set according to the comparison result and the difference data comparison tuple specifically includes:

determining the ordinate value of each position data; wherein, the ordinate axis of the position data is along the tangential direction perpendicular to the annular position where the auxiliary positioning sensor is located;

comparing each longitudinal coordinate value with the preset grouping threshold interval;

determining that the corresponding difference data comparison binary group is a pending group data binary group under the condition that the comparison result is that the ordinate value is not in the preset group threshold interval;

and respectively taking the moment in each group data binary group to be determined as the center, presetting a dividing time value as a neighborhood radius, determining a grouping time interval, determining positioning acquisition data corresponding to each grouping time interval, and taking the positioning acquisition data as the running difference data group.

In one implementation manner of the present application, determining the association degree value between each operation difference data set and the vibration signal based on the vibration signal from the bearing annular housing of the positioning detection device and a preset abnormal trigger association model specifically includes:

calculating the average value of each position data in the running difference data set as a first position average value through the abnormal trigger association model, and determining a first position standard deviation of each position data; and

calculating an average value of vibration intensity values of the vibration signals corresponding to the acquisition time of the position data, taking the average value as a first vibration intensity average value, and determining a first vibration intensity standard deviation corresponding to each vibration intensity value;

and inputting the first position mean value, the first position standard deviation, the first vibration intensity mean value and the first vibration intensity standard deviation into a Pearson correlation coefficient calculation formula of the abnormal trigger correlation model to determine the correlation degree value of each operation difference data set and the corresponding vibration signal according to a calculation result.

In one implementation manner of the present application, before determining the association degree value between each operation difference data set and the corresponding vibration signal according to the calculation result, the method further includes:

Determining a second position mean value and a second position standard deviation corresponding to the positioning acquisition data set through the abnormal triggering association model; and

determining a second vibration intensity mean value and a second vibration standard deviation of the vibration signals corresponding to the positioning acquisition data set;

calculating a first association coefficient of the positioning acquisition data set and the vibration signal through the Pearson correlation coefficient calculation formula;

and determining a second association coefficient of each operation difference data set and the corresponding vibration signal, and taking the product of the first association coefficient and the second association coefficient as the association degree value.

In one implementation manner of the present application, based on each of the correlation degree values and the vibration intensity values of the vibration signals, each of the operation difference data sets is subjected to data correction, so as to obtain each of the corrected operation data sets, which specifically includes:

according to each association degree value, matching association weights in a preset weight list;

performing cluster analysis on signal segments of the vibration signals corresponding to the operation difference data sets to obtain classified signal segment levels; the signal segmentation is obtained by dividing the vibration signal according to a data acquisition period corresponding to the operation difference data set;

Determining a historical operation difference data set corresponding to each signal segmentation level and a corresponding historical association weight correction value of each historical operation difference data set; wherein, the operation difference data of the history operation difference data group is obtained by vibration signals with different signal segmentation levels;

matching each operation difference data set with the corresponding historical operation difference data set through a preset matching rule so as to determine the historical association weight correction value corresponding to each operation difference data set;

and correcting the association weight according to the historical association weight correction value and a preset correction strategy, so as to respectively carry out data correction on each operation difference data set according to the corrected association weight, and obtain each corrected operation data set.

In one implementation manner of the present application, performing cluster analysis on signal segments of the vibration signal corresponding to each operation difference data set to obtain classified signal segment levels, which specifically includes:

determining signal segments of the vibration signals corresponding to the operation difference data sets, and generating vibration signal segment curves of the signal segments;

Calculating cosine similarity between the vibration signal segmentation curves through a preset cosine similarity formula, and generating a similarity matrix;

determining a signal similarity mean value based on the similarity matrix and the curve number of the vibration signal segmentation curves;

sequentially traversing the vibration signal segmentation curve group with the cosine similarity larger than the signal similarity mean value in the similarity matrix; the vibration signal segmentation curve group at least comprises two vibration signal segmentation curves;

clustering the vibration signal segmentation curves according to the vibration signal segmentation curve groups to obtain a plurality of clustering clusters;

and matching the signal segmentation level corresponding to each cluster according to the vibration signal segmentation curve and the historical vibration intensity classification list corresponding to the cluster center in the cluster.

In one implementation of the present application, the method further includes:

determining each operation data set corresponding to each auxiliary positioning sensor when the bearing retainer is provided with more than two auxiliary positioning sensors;

and according to the acquisition time of the corresponding positioning acquisition data, carrying out combination processing on each operation data set, so as to determine the corresponding bearing health state of the bearing retainer through the operation data set, the vibration signal and the corresponding monitoring and early warning standard after combination processing, and sending the bearing health state to a user terminal to monitor and early warn the operation of the fan bearing.

In an implementation manner of the present application, according to each operation data set, the vibration signal, and the corresponding monitoring and early warning standard, a corresponding bearing health state of the bearing retainer is determined, so that the bearing health state is sent to a user terminal, and the fan bearing operation is monitored and early warned, which specifically includes:

generating an operation positioning curve of the operation data set;

respectively determining the number of wave crest values, the wave crest value of which is larger than a first preset threshold value, and the number of wave trough values, the wave trough value of which is smaller than a second preset threshold value, in the operation positioning curve;

determining whether the vibration signal intensity of the vibration signal meets the monitoring and early warning standard;

if yes, matching corresponding bearing health states according to the number of the wave crests and the number of the wave troughs according to the monitoring and early warning standard; the bearing health state at least comprises life prediction data of the bearing retainer and bearing running condition data of the wind turbine generator;

otherwise, the monitoring and early warning standard is increased or decreased according to the vibration signal intensity and a preset threshold interval.

On the other hand, the embodiment of the application provides bearing operation monitoring equipment based on a wind power bearing retainer, and the equipment is applied to a wind turbine generator; the bearing retainer of the wind turbine generator is provided with at least one auxiliary positioning sensor for positioning the bearing retainer in real time and sending out corresponding positioning signals; the bearing annular shell is provided with a positioning detection device for collecting the positioning signals; the apparatus comprises:

At least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to:

based on the correlation degree values and the vibration intensity of the vibration signals, respectively carrying out data correction on each operation difference data set to obtain corrected operation data sets; one operation data set at least comprises one operation difference data set, and the corresponding association degree values of the operation difference data sets in the same operation data set are the same;

According to the scheme, the problems that the current bearing running state, the current bearing health state and other data are mainly obtained manually, information obtaining is delayed, and the real-time supervision of the fan bearing and the retainer is not realized are solved. According to the scheme, the radial displacement operation condition of the bearing retainer can be obtained through the positioning detection device and the auxiliary positioning sensor, the operation condition of the bearing retainer is corrected by combining the vibration intensity of the vibration signal, and then the health state of the bearing is obtained according to the corresponding monitoring and early warning standard. And further realizing the real-time supervision of the fan bearing and the retainer. When the bearing operates and monitors, on the premise of saving labor cost, the bearing operation data information is timely acquired, and the fan bearing and the retainer are monitored in real time.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:

FIG. 1 is a schematic flow chart of a method for monitoring bearing operation based on a wind power bearing retainer according to an embodiment of the present application;

FIG. 2 is a schematic annular cross-sectional view of a bearing cage in a method for monitoring operation of a bearing based on a wind power bearing cage according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a positioning monitoring device in a method for monitoring bearing operation based on a wind power bearing cage according to an embodiment of the present application;

fig. 4 is a schematic structural diagram of a bearing operation monitoring device based on a wind power bearing retainer in an embodiment of the application.

Detailed Description

For the purposes, technical solutions and advantages of the present application, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.

The wind turbine generator is generally in severe environments such as sufficient air quantity, high and cold, low temperature and the like, the bearing component of the wind turbine generator is an important component for wind power generation, and the requirements on the healthy service life of the bearing and the healthy running state of the bearing are high in the severe environments. Therefore, during the operation of the wind turbine generator, maintenance personnel are required to check the health condition of the bearing, such as the fatigue life of the bearing, the stable operation condition of the bearing and the like, and maintain the bearing regularly.

On one hand, the method for manually monitoring the bearing operation may need to perform shutdown monitoring, so that the manual detection time is long, the monitoring efficiency is low, and the purposes of timely acquiring monitoring data and timely maintaining cannot be achieved; on the other hand, the number of the wind turbine generators is large, the detection period for manually detecting the bearings of each wind turbine generator one by one is long, and the data information acquisition is delayed. If the operation and the power generation efficiency of the wind turbine generator can be improved, the labor cost can be reduced, and the healthy operation state of the wind turbine generator can be ensured.

Based on the above, the embodiment of the application provides a bearing operation monitoring method and device based on a wind power bearing retainer, which are used for solving the problems that the current bearing operation state, health condition and other data are mainly acquired manually, information acquisition is delayed, and real-time supervision of a fan bearing and the retainer is not realized.

Various embodiments of the present application are described in detail below with reference to the accompanying drawings.

The embodiment of the application provides a bearing operation monitoring method based on a wind power bearing retainer, which is applied to a wind turbine generator. The bearing retainer of the wind turbine generator is provided with at least one auxiliary positioning sensor for positioning the bearing retainer in real time and sending out corresponding positioning signals. The bearing annular shell is provided with a positioning detection device for collecting positioning signals. As shown in fig. 1, the method may include steps S101-S105:

S101, a server acquires a positioning acquisition data set from a positioning detection device.

Wherein the positioning acquisition data set at least comprises position data of the auxiliary positioning sensor. The position data refers to position coordinates, such as the coordinate system in fig. 2, in which the position data of the auxiliary positioning sensor is coordinates.

The auxiliary positioning sensor may be disposed on the bearing holder, as shown in fig. 2, where the auxiliary positioning sensors such as 201, 202, 203, 204 are disposed on the annular cross section side of the bearing holder, and if the position of the coordinate system set in fig. 2 is taken as the initial position of the bearing holder for rotation, the position data of the auxiliary positioning sensor at the initial position is standard position data; and a positioning detection device is arranged on the bearing annular shell and positioned right above the auxiliary positioning sensor 201. The positioning signals sent by the auxiliary positioning sensor can be collected. In the practical use process, the application can adopt any one or a plurality of auxiliary positioning sensors and any number of positioning detection devices, and only when the bearing retainer is produced, the center of mass of the bearing retainer after the auxiliary positioning sensors are installed is ensured to be centered.

The auxiliary positioning sensor can send out a positioning signal (such as a laser signal and a near field communication signal (Near Field Communication, NFC)) on the horizontal plane where the auxiliary positioning sensor is located, and correspondingly the positioning detection device can collect the horizontal plane where the positioning signal is located. As shown in fig. 3, the positioning detection device is configured to collect the positioning signal of the auxiliary positioning sensor, and the signal collecting surface 310 and the connector 320 are provided with a signal collecting array, such as including

signal collecting areas

311, 312, 313, where in actual use, more areas may be included, and each area of the signal collecting array can be used for collecting the positioning signal and separately transmitting the positioning signal or separately transmitting the strength of the collected positioning signal. The signal collection surface 310 is parallel to the annular cross section of the bearing cage, and when the bearing cage is set in its initial position, the signal collection area 312 in the signal collection surface 310 and the auxiliary positioning sensor can be set in the same horizontal plane and opposite. The connector 320 is fixedly connected to the bearing annular housing.

In an embodiment of the present application, if the positioning detection device collects positioning signals, when only one signal collection area can collect positioning signals (such as laser) at the same time, the server may match position data corresponding to the signal collection area according to the signal collection area where the positioning signals are collected. And taking the position data as the position data of the auxiliary positioning sensor, further obtaining the position data of the auxiliary positioning sensor when the auxiliary positioning sensor passes through the signal acquisition area each time, and generating a positioning acquisition data set. The server stores position data of positions set by the signal acquisition areas in advance.

In another embodiment of the present application, if the positioning detection device collects positioning signal intensity, there may be a plurality of signal collection areas that all collect positioning signals, but there is a difference between the signal intensities collected by the signal collection areas, and the server selects a signal collection area with the largest signal intensity, and uses the position data corresponding to the signal collection area as the position data of the positioning signal sent by the auxiliary positioning sensor. If the maximum signal intensity corresponds to more than one signal acquisition area, the server can take the average value of the position data corresponding to the two signal acquisition areas as the position data of the positioning signals sent by the auxiliary positioning sensor.

The positioning acquisition data set may be position data acquired in real time, or may be position data obtained by screening according to a preset acquisition time interval, which is not particularly limited in this application.

The server is an execution subject of the bearing operation monitoring method based on the wind power bearing holder, and the execution subject is not limited to the server, but is not particularly limited thereto.

S102, the server analyzes the positioning acquisition data set according to the data acquisition time sequence by presetting a difference data positioning algorithm so as to determine an operation difference data set of the bearing retainer.

The position data in the same operation difference data group are the same as the difference value of the preset standard position data.

In this embodiment of the present application, by a preset differential data positioning algorithm, according to a data acquisition time sequence, a positioning acquisition data set is analyzed to determine an operation differential data set of a bearing retainer, which specifically includes:

firstly, the server divides data of a positioning acquisition data set in the same monitoring period into a plurality of operation monitoring analysis sequences according to the monitoring period divided in advance.

The monitoring time period is obtained by dividing the preset environmental condition change time period. Each operational monitoring analysis sequence includes a plurality of positioning acquisition data. The positioning acquisition data at least comprises a binary group consisting of time and position data.

The monitoring period can be divided by a user according to the environmental condition change period, for example, under certain environmental conditions, the wind turbine generator needs to stop running and overhauling, and monitoring is not needed; under certain environmental conditions, the monitoring and early warning standard for monitoring the bearing carrier is special, and a user can divide the duration corresponding to the environmental conditions into a single monitoring period to match the specific monitoring and early warning standard; and for example, the monitoring period is only aimed at certain environmental conditions such as strong wind, rain, etc. The specific dividing mode of the monitoring time period is not particularly limited, the calculated amount of the server during single data processing can be reduced by dividing the monitoring time period, and the bearing operation supervision efficiency is ensured.

The server may divide the positioning acquisition data in the positioning acquisition data set in the same monitoring period into a plurality of operation monitoring analysis sequences, and the number of operation monitoring analysis sequences may be a plurality, and specific values of the plurality may be set by a user, for example, the setting mode is any number between a-b, etc.

And then, the server divides each positioning acquisition data in each operation monitoring analysis sequence into a plurality of operation monitoring analysis subsequences through a difference data positioning algorithm, and sequentially determines the median of each operation monitoring analysis subsequence according to a preset rule. The median is obtained by comparing the position data in the tuples.

The difference data positioning algorithm in the application can be a linear search algorithm, the server can randomly divide the operation monitoring analysis sequence into a plurality of operation monitoring analysis subsequences, the operation monitoring analysis sequence comprises a plurality of positioning acquisition data, and the application can generate a plurality of operation monitoring analysis subsequences in a mode that 5 positioning acquisition data form a sequence or 7 positioning acquisition data form. And determining the median in each run monitor analysis sub-sequence. The median is in fact the median of the position data, i.e. the coordinate values, more precisely the ordinate values, in the positioning acquisition data. Because the moments are aligned linearly, and the position data in the run monitor analysis sequence is not, the bearing cage may be displaced by a different amount in the radial direction at each moment, and the position data may be different at each moment.

The server then generates a corresponding median sequence for each median by a bubble ordering algorithm to determine a selected median in the median sequence.

The bubble ordering algorithm can sequentially compare two adjacent elements, and move and arrange the elements in the sequence according to the size relation of the compared elements to obtain the sequence from large to small or from small to large. The server can arrange the median into a median sequence by the bubble ordering algorithm, and can determine the median in the median sequence as a selected median. In determining the median, if there are two medians, any one may be selected as the median of the sequence. For example, the median sequence is [1,2,3,4,5,6,7,8,9], the median is selected to be 5, and further, for example, the median sequence is [1,2,3,4,5,6,7,8], the median is selected to be 4.

Then, the server determines the difference data comparison binary group of the element positions corresponding to the difference data screening values in each operation monitoring analysis sequence based on the selected median, the operation monitoring analysis sequence and the preset difference data screening values. The difference data comparison binary group comprises position data and corresponding time.

After the server obtains the selected median, the running monitoring analysis sequence can be divided by the selected median, and the sub-sequence A is used as a dividing sub-sequence greater than the selected median and the sub-sequence B is used as a dividing sub-sequence less than or equal to the selected median. The server can determine the element number value B in the divided subsequence B, compare the magnitude relation between B and the difference data screening value c, and when the difference data screening value c is smaller than or equal to B, search the data with the element position ordering sequence number equal to the difference data screening value c from the divided subsequence as the difference data comparison binary group. The difference data comparison binary set includes: time of day, location data. And when the difference data screening value c is larger than b, searching the positioning acquisition data corresponding to the c-b element position in the dividing subsequence A.

The preset difference data screening value is set by a user, and aims to ensure that when the difference data is selected to be compared with the binary groups, the longitudinal coordinate values of the same numerical value size arrangement position are selected from each operation monitoring analysis sequence so as to ensure that the selected positioning acquisition data are uniform. The same value size arrangement position refers to that for example, N elements are arranged in a sequence, and elements with the c-th largest size sequence can be selected through difference data screening value c; in another sequence, there are N elements, and the c-th element can be selected. Through the scheme, the position data with large values can be screened from each operation monitoring analysis sequence by the first difference data, so that each difference data comparison binary set is obtained, the condition that the maximum or minimum value and the like in the sequence are selected in the selected data, no representative data influence the referenceable value of the subsequently generated operation difference data set, and finally the bearing operation monitoring effect is influenced is avoided.

And finally, the server compares the position data in each difference data comparison binary group with a preset grouping threshold interval so as to generate an operation difference data group according to the comparison result and the difference data comparison binary group.

The preset grouping threshold interval can be set by a user, or can be determined by referring to an average value of the radial running displacement of the historical bearing retainer and adding a neighborhood error, wherein the neighborhood error refers to an error in a part of running process. For example, the average value is 0, the field error is 0.5, the preset grouping threshold interval is [ -0.5,0.5], the numerical values are only examples, and the preset grouping threshold interval is not particularly limited in the application.

The method comprises the steps of comparing position data in each difference data comparison binary group with a preset grouping threshold interval to generate an operation difference data group according to comparison results and the difference data comparison binary group, and specifically comprises the following steps:

the server determines the ordinate values of the respective position data. Wherein the ordinate axis of the position data is along a tangential direction perpendicular to the annular position where the auxiliary positioning sensor is located. And comparing each ordinate value with a preset grouping threshold interval. And under the condition that the comparison result is that the ordinate value is not in the preset grouping threshold interval, determining that the corresponding difference data comparison binary group is a pending grouping data binary group. And respectively taking the time in each group of undetermined group data as the center, presetting a dividing time value as a neighborhood radius, determining a group time interval, determining positioning acquisition data corresponding to each group time interval, and taking the positioning acquisition data as an operation difference data group.

The preset dividing time value is, for example, 10 seconds, 5 seconds, etc., the grouping time interval is, for example, [1, 20], the number unit may be seconds, or may be set to be minutes, and the number unit of the dividing time value is not specifically limited in this application.

The operation difference data set is obtained by dividing the obtained difference data comparison binary groups, namely, the operation difference data set is divided according to the data that the radial displacement of the bearing retainer is not in the preset grouping threshold interval, and the operation difference data set with overlarge displacement corresponding to different time can be obtained.

In addition, in the running process of the wind turbine generator system, the situation that the bearing retainer slips is likely to be faced, and the data of the obtained running difference data set caused by the slipping of the bearing retainer can be avoided through the following technical scheme.

The server can determine the vibration intensity value of the vibration signal at each acquisition time, the running time of the bearing retainer, the life prediction abrasion value preset by a user, and a calculation formula according to the slip risk

Slip risk value->

，/>

Is a preset long-time weight of use, +.>

Weight of vibration intensity value of vibration signal, +.>

Predicting a wear value weight for a preset lifetime, +. >

For the duration of use, +.>

For the vibration intensity value, < >>

Wear values are predicted for lifetime. The slip risk calculation formula can be obtained by adjusting weight parameters according to experience in the actual use process of a user. According to the slip risk corresponding to each acquisition time, when the slip risk is larger than a preset risk threshold set by a user, the corresponding slip time period adjustment step length can be matched, the acquisition time in the positioning acquisition data set can be adjusted by the slip time period adjustment step length, for example, the slip time period adjustment step length is 00:01, and two data acquisition times in the positioning acquisition data set are 00:05, and 00:08, the server adjusts the two data acquisition moments to 00:04, 00:06.

S103, the server determines the association degree value of each operation difference data set and the vibration signal based on the vibration signal of the bearing annular shell from the positioning detection device and a preset abnormal trigger association model.

The method for determining the association degree value of each operation difference data set and the vibration signal based on the vibration signal of the bearing annular shell from the positioning detection device and a preset abnormal trigger association model specifically comprises the following steps:

the server calculates the average value of the position data in the running difference data set through the abnormal triggering association model to be a first position average value, and determines a first position standard deviation of the position data. And calculating an average value of vibration intensity values of the vibration signals corresponding to the acquisition time of the position data as a first vibration intensity average value, and determining a first vibration intensity standard deviation corresponding to each vibration intensity value.

And then, inputting the first position mean value, the first position standard deviation, the first vibration intensity mean value and the first vibration intensity standard deviation into a pearson correlation coefficient calculation formula of an abnormal trigger correlation model, so as to determine the correlation degree value of each operation difference data set and the corresponding vibration signal according to the calculation result.

That is, the present application may calculate the average value of the position data (ordinate value) in each operation difference data group through the abnormal trigger correlation model, respectively, to obtain the first position average value. And calculating a first position standard deviation according to the first position mean and the position data. Similarly, a first vibration intensity mean value and a first vibration intensity standard deviation of a vibration intensity value of the vibration signal are calculated, and then a pearson correlation coefficient calculation formula is utilized, wherein the specific formula is as follows:

wherein,,

for the pearson correlation coefficient, +.>

For the number of data in the running disparity data set, < +.>

For the position data in the running disparity data set, +.>

For the first location mean>

For the first position standard deviation, +.>

For the vibration intensity value of the vibration signal, +.>

Is the first vibration intensity mean value->

Is the first vibration intensity standard deviation.

Through the abnormal triggering association model, the association relation between the operation difference data set and the vibration signal corresponding to the operation difference data set can be established, and the association degree value of the association relation is calculated.

Furthermore, before determining the association degree value of each operation difference data set and the corresponding vibration signal according to the calculation result, the method further comprises:

and the server determines a second position mean value and a second position standard deviation corresponding to the positioning acquisition data set through the abnormal triggering association model. And determining a second vibration intensity mean value and a second vibration standard deviation of the vibration signals corresponding to the positioning acquisition data set. And calculating a first correlation coefficient of the positioning acquisition data set and the vibration signal through a Pearson correlation coefficient calculation formula. And determining a second correlation coefficient of each operation difference data set and the corresponding vibration signal, and taking the product of the first correlation coefficient and the second correlation coefficient as a correlation degree value.

That is, the method and the device can calculate the association relation between the positioning acquisition data set and the vibration signal before grouping, determine the first association coefficient between the positioning acquisition data set and the vibration signal on the whole, determine each second association coefficient from the grouping, obtain the association degree value according to the association coefficient between the whole and the individual, and further enable the association degree value between the position data and the vibration signal to be more accurate and reliable.

And S104, the server respectively corrects the data of each operation difference data set based on each association degree value and the vibration intensity value of the vibration signal so as to obtain each corrected operation data set.

The operation data set at least comprises an operation difference data set, and the corresponding association degree values of the operation difference data sets in the same operation data set are the same.

In this embodiment of the present application, based on each correlation degree value and the vibration intensity value of the vibration signal, data correction is performed on each operation difference data set, so as to obtain each corrected operation data set, which specifically includes:

firstly, the server matches the association weight in a preset weight list according to each association degree value.

The preset weight list is set by the user, because the association degree value is between [ -1,1] and is inconvenient to calculate, the association weight corresponding to the association degree value can be set, and the weight list is obtained, and the weight list can be set by referring to historical data in the actual use process, which is not particularly limited in the application.

Then, the server performs cluster analysis on the signal segments of the vibration signals corresponding to each operation difference data set to obtain the classified signal segment levels.

The signal segmentation is obtained by dividing the vibration signal according to the data acquisition time period corresponding to the operation difference data set.

Specifically, the server may determine signal segments of the vibration signal corresponding to each of the running difference data sets, and generate a vibration signal segment curve for each of the signal segments. And calculating cosine similarity between the vibration signal segmentation curves through a preset cosine similarity formula, and generating a similarity matrix. And determining a signal similarity mean value based on the similarity matrix and the curve number of the vibration signal segmentation curves.

That is, the server may perform signal segmentation on the vibration signal according to the time segmentation of the operation difference data set after the above-described grouping, and generate a vibration signal segmentation curve for each signal segment. Then, the server can calculate cosine similarity between curves through a cosine similarity calculation formula, and a similarity matrix with the cosine similarity between the curves is established. Then, the server can determine the signal similarity mean value according to the cosine similarity and the curve number in the similarity matrix.

The specific calculation mode of the signal similarity mean value may be that the server calculates a sum value of cosine similarity in the similarity matrix, and divides the sum value by the total curve number to obtain the signal similarity mean value.

The server sequentially traverses vibration signal segmentation curve groups with cosine similarity larger than the average value of the signal similarity in the similarity matrix. The vibration signal segmentation curve group at least comprises two vibration signal segmentation curves. And then, clustering the vibration signal segmentation curves according to the vibration signal segmentation curve groups to obtain a plurality of clustering clusters. And matching the signal segmentation level corresponding to each cluster according to the vibration signal segmentation curve and the historical vibration intensity classification list corresponding to the cluster center in the cluster.

And each cosine similarity corresponds to two vibration signal segmentation curves, and the server adds the vibration signal segmentation curves with the cosine similarity larger than the average value of the signal similarity to the same curve group to obtain the vibration signal segmentation curve group. For example, starting with the first vibration signal segmentation curve, traversing, and adding the vibration signal segmentation curves with cosine similarity larger than the average value of the signal similarity with the first vibration signal segmentation curve into the first vibration signal segmentation curve group. And then starting to traverse by using the second vibration signal segmentation curve, wherein the traversing is completed to generate a curve group, and the subsequent traversing process is not participated, namely the second vibration signal segmentation curve is traversed, and the cosine similarity between the second vibration signal segmentation curve and the first vibration signal segmentation curve is not compared. The server generates each vibration signal segment curve group, and then clusters the vibration signal segment curve groups. If the cosine similarity of at least one curve of the vibration signal segmentation curve group M1 and a plurality of curves of the vibration signal segmentation curve group M2 is larger than the signal similarity mean value, combining the two vibration signal segmentation curves until all the vibration signal segmentation curve groups are traversed, and obtaining a plurality of clustering clusters.

Then, the server can determine a cluster center in each cluster, and match the vibration signal segmentation curve corresponding to the cluster center with a preset historical vibration intensity classification list, which can be set by the user, and the application is not limited in particular. The corresponding relation between a plurality of preset curves and preset signal segmentation levels can be recorded in the historical vibration intensity classification list, the server can calculate the similarity between the vibration signal segmentation curves and each preset curve, for example, the friendship distance is taken as the similarity between the two curves, the signal segmentation level corresponding to the preset curve with the minimum friendship distance is selected as the signal segmentation level of the clustering center, the signal segmentation level corresponding to each clustering cluster is further determined, the signal segmentation level can be set by a user, and the application is not limited in particular. The present application may also determine the curve similarity by other means, which is not specifically limited in this application.

Then, the server determines a historical operation difference data set corresponding to each signal segment level and a corresponding historical association weight correction value of each historical operation difference data set. Wherein, the operation difference data of the historical operation difference data group is obtained by vibration signals with different signal segmentation levels. And matching each operation difference data set with the corresponding historical operation difference data set through a preset matching rule so as to determine a historical association weight correction value corresponding to each operation difference data set.

After obtaining the signal segment level of each cluster, the server can determine a historical operation difference data set corresponding to the signal segment level, wherein the historical operation difference data set can be an operation difference data set before the current operation difference data set is acquired or in the same monitoring period before. The historical operation difference data sets are provided with historical association weight correction values, and the historical operation difference data sets are matched through the determined signal segmentation level, so that the historical association weight correction values corresponding to the operation difference data sets are obtained. The preset matching rule may be that data in the two data sets are sequentially and correspondingly calculated according to an euclidean distance average value, and one historical operation difference data set with the smallest euclidean distance average value is used as the matched historical operation difference data set, so as to obtain the historical association weight correction value.

And finally, the server corrects the association weight according to the historical association weight correction value and a preset correction strategy, so that the data correction is respectively carried out on each operation difference data set according to the corrected association weight, and each corrected operation data set is obtained.

The historical associated weight correction value is a correction value for eliminating noise of the associated weight, and the preset correction strategy can understand that the correction value can be added or subtracted for the associated weights with different sizes, so as to correct the associated weights. And multiplying the corrected association weight with the position data in the operation difference data set respectively, and correcting the position data to obtain a corrected operation data set.

S105, the server determines the corresponding bearing health state of the bearing retainer according to each operation data set, the vibration signal and the corresponding monitoring and early warning standard so as to send the bearing health state to the user terminal and monitor and early warn the operation of the fan bearing.

The user terminal may be a mobile phone, a computer or other equipment of a wind turbine generator supervisory personnel, which is not particularly limited in this application. The server sends the health state of the bearing to the user terminal, and the user can check the health state of the bearing and take corresponding maintenance measures.

In the embodiment of the present application, in the case where the bearing holder is provided with two or more auxiliary positioning sensors, each operation data set corresponding to each auxiliary positioning sensor is determined. And according to the corresponding acquisition time of positioning acquisition data, carrying out combination processing on each operation data set, so as to determine the corresponding bearing health state of the bearing retainer through the operation data set after combination processing, the vibration signal and the corresponding monitoring and early warning standard, and sending the bearing health state to a user terminal to monitor and early warn the operation of the fan bearing.

That is, as in fig. 2, when there are multiple auxiliary positioning sensors, the server may combine the operation data sets corresponding to the auxiliary positioning sensors into the same operation data set according to time, and then perform monitoring and early warning on the operation of the fan bearing, which can avoid the error of collecting data for only one auxiliary positioning sensor, and improve the accuracy of monitoring and early warning.

In this embodiment of the application, according to each operation data group and vibration signal, corresponding monitoring early warning standard, confirm the corresponding bearing health status of bearing holder to send bearing health status to user terminal, monitor early warning to fan bearing operation, specifically include:

the server may generate an operational location curve for the operational data set. And respectively determining the wave crest value quantity of which the wave crest value is larger than a first preset threshold value and the wave trough value quantity of which the wave trough value is smaller than a second preset threshold value in the operation positioning curve. And determining whether the vibration signal intensity of the vibration signal meets the monitoring and early warning standard. And under the condition that the vibration signal intensity of the vibration signal is determined to be in accordance with the monitoring and early warning standard, matching the corresponding health state of the bearing according to the monitoring and early warning standard and the number of wave peak values and wave trough values. The bearing health state at least comprises life prediction data of the bearing retainer and bearing running condition data of the wind turbine generator. Otherwise, the monitoring and early warning standard is improved or reduced according to the vibration signal intensity and a preset threshold interval.

The first preset threshold and the second preset threshold are set by the user, which is not particularly limited in this application. When the monitoring and early warning standard such as vibration signal intensity is in a certain threshold value interval of the monitoring and early warning standard, according to the peak value data and the trough value quantity, bearing health state information matched with the monitoring and early warning standard such as x-time reduction of bearing life prediction data, poor bearing running conditions and the like is generated. The monitoring and early warning standard is provided with a plurality of monitoring and early warning threshold intervals, and if the current vibration signal strength does not meet the currently selected monitoring and early warning marked threshold interval, the server can adjust the monitoring and early warning standard so as to be matched with the corresponding bearing health state.

According to the scheme, the radial displacement operation condition of the bearing retainer can be obtained through the positioning detection device and the auxiliary positioning sensor, the vibration intensity of the vibration signal is combined, the operation condition of the bearing retainer is corrected, and then the health state of the bearing is obtained according to the corresponding monitoring and early warning standard. And further realizing the real-time supervision of the fan bearing and the retainer. When the bearing operates and monitors, on the premise of saving labor cost, the bearing operation data information is timely acquired, and the fan bearing and the retainer are monitored in real time. The problems that the current data such as the running state and the health condition of the bearing mainly depend on manual acquisition, information acquisition is delayed and real-time supervision of the fan bearing and the retainer is not realized are solved.

Fig. 4 is a schematic structural diagram of a bearing operation monitoring device based on a wind power bearing retainer, which is provided in an embodiment of the present application, and the device is applied to a wind turbine generator. The bearing retainer of the wind turbine generator is provided with at least one auxiliary positioning sensor for positioning the bearing retainer in real time and sending out corresponding positioning signals. The bearing annular shell is provided with a positioning detection device for collecting positioning signals. The device comprises:

At least one processor; and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to:

a positioning acquisition dataset from a positioning detection device is acquired. Wherein the positioning acquisition data set at least comprises position data of the auxiliary positioning sensor. And analyzing the positioning acquisition data set according to the data acquisition time sequence by a preset difference data positioning algorithm to determine an operation difference data set of the bearing retainer. The position data in the same operation difference data group are the same as the difference value of the preset standard position data. And determining the association degree value of each operation difference data set and the vibration signal based on the vibration signal of the bearing annular shell from the positioning detection device and a preset abnormal trigger association model. And respectively carrying out data correction on each operation difference data set based on each association degree value and the vibration intensity of the vibration signal so as to obtain each corrected operation data set. One operation data set at least comprises one operation difference data set, and the corresponding association degree values of the operation difference data sets in the same operation data set are the same. And determining the corresponding bearing health state of the bearing retainer according to each operation data set, the vibration signal and the corresponding monitoring and early warning standard so as to send the bearing health state to the user terminal and monitor and early warn the operation of the fan bearing.

All embodiments in the application are described in a progressive manner, and identical and similar parts of all embodiments are mutually referred, so that each embodiment mainly describes differences from other embodiments. In particular, for the apparatus embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.

The devices and the methods provided in the embodiments of the present application are in one-to-one correspondence, so that the devices also have similar beneficial technical effects as the corresponding methods, and since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices are not described here again.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. The bearing operation monitoring method based on the wind power bearing retainer is characterized by being applied to a wind turbine generator; the bearing retainer of the wind turbine generator is provided with at least one auxiliary positioning sensor for positioning the bearing retainer in real time and sending out corresponding positioning signals; the bearing annular shell is provided with a positioning detection device for collecting the positioning signals; the method comprises the following steps:

2. The method for monitoring the operation of a bearing based on a wind power bearing retainer according to claim 1, wherein the positioning acquisition data set is analyzed according to a data acquisition time sequence by a preset difference data positioning algorithm to determine an operation difference data set of the bearing retainer, specifically comprising:

and comparing the position data in each difference data comparison binary group with a preset grouping threshold interval to generate the operation difference data group according to a comparison result and the difference data comparison binary group.

3. The method for monitoring bearing operation based on a wind power bearing retainer according to claim 2, wherein the comparing the position data in each of the differential data comparison tuples with a preset grouping threshold interval to generate the operation differential data set according to the comparison result and the differential data comparison tuple, specifically comprises:

4. The method for monitoring bearing operation based on a wind power bearing retainer according to claim 1, wherein determining the association degree value of each operation difference data set and the vibration signal based on the vibration signal of the bearing annular housing from the positioning detection device and a preset abnormal trigger association model specifically comprises:

5. The method for monitoring bearing operation based on a wind power bearing retainer according to claim 4, wherein before determining the correlation degree value of each operation difference data set and the corresponding vibration signal according to the calculation result, the method further comprises:

6. The method for monitoring bearing operation based on a wind power bearing retainer according to claim 1, wherein each operation difference data set is subjected to data correction based on each correlation degree value and vibration intensity value of the vibration signal, respectively, so as to obtain each corrected operation data set, specifically comprising:

7. The method for monitoring bearing operation based on a wind power bearing retainer according to claim 6, wherein the step of performing cluster analysis on the signal segments of the vibration signal corresponding to each operation difference data set to obtain the classified signal segment levels comprises the steps of:

8. A method of monitoring bearing operation based on a wind power bearing cage according to claim 1, further comprising:

9. The method for monitoring the operation of the bearing based on the wind power bearing retainer according to claim 8, wherein the corresponding health state of the bearing retainer is determined according to each operation data set, the vibration signal and the corresponding monitoring and early warning standard, so as to send the health state of the bearing to a user terminal, and the operation of the bearing of the fan is monitored and early warned, specifically comprising:

generating an operation positioning curve of the operation data set;

10. Bearing operation monitoring equipment based on wind power bearing retainer, which is characterized in that the equipment is applied to a wind turbine generator; the bearing retainer of the wind turbine generator is provided with at least one auxiliary positioning sensor for positioning the bearing retainer in real time and sending out corresponding positioning signals; the bearing annular shell is provided with a positioning detection device for collecting the positioning signals; the apparatus comprises:

at least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein,,