Bearing state monitoring parameter screening method and bearing state monitoring method

Abstract

The invention provides a bearing state monitoring parameter screening method and a bearing state monitoring method, and relates to the technical field of bearing monitoring. Through the bearing state monitoring parameter screening method, the dimensionless statistical characteristic parameters capable of more accurately representing the vibration state of the bearing can be screened out for bearing state monitoring representation, so that early defects of the bearing can be identified. And meanwhile, the statistical characteristic parameters capable of eliminating data interception length interference obtained by screening through the bearing state monitoring parameter screening method serve as the bearing state monitoring parameters for bearing vibration characteristic characterization, the confidence coefficient is higher, and therefore the bearing state can be effectively and accurately characterized.

Description

Technical Field

[0001] The present invention relates to the technical field of bearing monitoring, and in particular to a bearing state monitoring parameter screening method and a bearing state monitoring method. Background Art

[0002] Aircraft bearings are generally installed between the rotating shaft and the casing of the aircraft tester through an interference fit. In order to meet the vibration measurement requirements of the bearing state, the vibration sensor is generally installed on the outside of the casing on the bearing force transmission path. The vibration impact caused by the bearing defect will be captured by the sensor. The allowable bearing defects of aircraft bearings are small, and it is very important to accurately capture the characteristics of early bearing defects. It is difficult to analyze early bearing defects directly using vibration amplitude and characteristic frequency analysis, and some dimensionless statistical values ​​are highly sensitive to defect impacts. Moreover, the dimensionless statistical value is independent of the bearing size parameters and can directly reflect the defect size, which is more suitable for the identification and evaluation of early bearing defects.

[0003] However, the current monitoring accuracy of bearing condition is not high, which can easily lead to large errors in the accurate identification of early bearing defects, thus affecting the timely and effective monitoring of bearing defects. Summary of the invention

[0004] The purpose of the present invention is to provide a bearing condition monitoring parameter screening method and a bearing condition monitoring method, which can monitor the bearing condition more accurately and help identify early defects of the bearing.

[0005] The embodiments of the present invention can be implemented in the following ways:

[0006] A method for screening bearing condition monitoring parameters, comprising: determining a dimensionless statistical parameter target for bearing condition monitoring; wherein the dimensionless statistical parameter target comprises a plurality of statistical characteristic parameters;

[0007] Acquire a data set; wherein the data set includes historical bearing vibration signal data;

[0008] According to the screening data, a bearing condition monitoring parameter for bearing vibration feature characterization is obtained in the dimensionless statistical parameter target; wherein the screening data is historical bearing vibration signal data selected from the data set for screening; the screening process includes:

[0009] Segmenting the screening data according to different data lengths, and calculating the time-domain statistical root mean square values ​​corresponding to the multiple statistical characteristic parameters according to the segmented screening data;

[0010] If the time-domain statistical RMS value satisfies the first preset condition, the statistical characteristic parameter corresponding to the time-domain statistical RMS value passes the screening.

[0011] Optionally, the step of obtaining bearing condition monitoring parameters that can be used to characterize bearing vibration characteristics in the dimensionless statistical parameter target based on the screening data includes multi-level screening; each level of screening selects the historical bearing vibration signal data with preset characteristics from the data set as the screening data and executes the screening process; and uses the statistical characteristic parameters that pass the screening in the last level of screening as the bearing condition monitoring parameters.

[0012] Optionally, the multi-level screening includes a first-level screening; the preset feature of the first-level screening is having any defects; the first-level screening includes:

[0013] The plurality of statistical characteristic parameters are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as primary screening results.

[0014] Optionally, the multi-level screening further includes a second-level screening; the preset feature of the second-level screening is a preset defect type; the second-level screening includes:

[0015] The first-level screening results are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as the second-level screening results.

[0016] Optionally, the multi-level screening further includes a third-level screening; the preset characteristics of the third-level screening are having the preset defect type and having a preset defect degree; the third-level screening includes:

[0017] The secondary screening results are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as the tertiary screening results.

[0018] Optionally, the multi-level screening further includes a fourth-level screening; the preset characteristics of the fourth-level screening are having the preset defect type and the preset defect degree and having a preset rotation speed; the fourth-level screening includes:

[0019] The third-level screening results are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as the fourth-level screening results.

[0020] Optionally, the first preset condition includes: the time domain statistical root mean square value calculated under different data clipping lengths is stable, or there is a data length clipping range in which the time domain statistical root mean square value is stable.

[0021] Optionally, the time-domain statistical root mean square value being stable means that when the data interception length increases by 100, the change of the time-domain statistical root mean square value does not exceed 10%.

[0022] Optionally, the screening data is segmented, and the step of calculating the time-domain statistical root mean square values ​​corresponding to the multiple statistical characteristic parameters according to each segment of data includes:

[0023] The screening data is divided into m segments of equal length by length L; each data segment of length L contains n data points;

[0024] Calculating the time domain statistical parameter value corresponding to the statistical characteristic parameter according to the n data points in each data segment to obtain m time domain statistical parameter values;

[0025] A root mean square value is calculated according to the m time-domain statistical parameter values ​​to obtain the time-domain statistical root mean square value.

[0026] Optionally, the plurality of statistical characteristic parameters include a kurtosis factor K u , volatility factor C r , Pulse Factor I m , marginal factor P r and the form factor W a at least part of .

[0027] Optionally, the formula for calculating the time domain statistical parameter value corresponding to the statistical feature parameter according to n data points in each data segment is as follows:

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036]

[0037] Where j represents the jth data segment; x(t) is the number of n discrete data points {x1, x2, x3…x n} constitutes a one-dimensional array;

[0038] The time domain statistical parameter values ​​corresponding to the m data segments can be expressed as:

[0039]

[0040] Optionally, the step of acquiring a bearing condition monitoring parameter for bearing vibration feature characterization in the dimensionless statistical parameter target according to the screening data comprises:

[0041] Primary screening: the primary screening is performed through the screening process to obtain a primary screening result;

[0042] Secondary screening: select historical bearing vibration signal data with a preset defect size as screening data; determine whether the relationship between the preset defect size corresponding to the screening data and the time domain statistical root mean square value corresponding to the statistical characteristic parameter in the first screening result meets the second preset condition; if the second preset condition is met, use the statistical characteristic parameter corresponding to the time domain statistical root mean square value as the bearing state monitoring parameter.

[0043] A bearing condition monitoring method, the bearing condition monitoring method comprising:

[0044] Acquire bearing state monitoring parameters for bearing vibration characteristic characterization; wherein the bearing state monitoring parameters are obtained according to the bearing state monitoring parameter screening method as described above;

[0045] The bearing state is monitored according to the bearing state monitoring parameters.

[0046] The beneficial effects of the bearing condition monitoring parameter screening method and the bearing condition monitoring method provided by the embodiments of the present invention include:

[0047] An embodiment of the present invention provides a bearing state monitoring parameter screening method, which includes: determining a dimensionless statistical parameter target for bearing state monitoring; wherein the dimensionless statistical parameter target includes multiple statistical characteristic parameters; obtaining a data set; wherein the data set includes historical bearing vibration signal data; selecting historical bearing vibration signal data for screening from the data set as screening data; obtaining a bearing state monitoring parameter for bearing vibration feature characterization in the dimensionless statistical parameter target according to the screening data; wherein the screening process includes: segmenting the screening data with different data lengths, and calculating the time domain statistical root mean square values ​​corresponding to multiple statistical characteristic parameters according to the segmented screening data; if the time domain statistical root mean square value meets the first preset condition, the statistical characteristic parameter corresponding to the time domain statistical parameter is used as the bearing state monitoring parameter. The bearing state monitoring parameter screening method selects statistical characteristic parameters that can eliminate the interference of data interception length as bearing state monitoring parameters for bearing vibration feature characterization, with higher confidence, so that the bearing state can be effectively and accurately characterized, which is helpful for identifying early defects of the bearing.

[0048] An embodiment of the present invention also provides a bearing state monitoring method, which characterizes the bearing state through the bearing state monitoring parameters screened out by the above-mentioned bearing state monitoring parameter screening method. Therefore, it also has the beneficial effect of being able to effectively and accurately characterize the bearing state and helping to identify early defects of the bearing. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The above features and advantages of the present invention can be better understood after reading the detailed description of the embodiments of the present disclosure in conjunction with the following drawings. In the drawings, the components are not necessarily drawn to scale, and components with similar related properties or features may have the same or similar reference numerals.

[0050] Figure 1 A flow chart showing a method for screening bearing condition monitoring parameters according to one aspect of the present invention is shown;

[0051] Figure 2 A schematic diagram showing the execution of step S311 by taking the kurtosis factor as an example according to one aspect of the present invention is shown;

[0052] Figure 3 A schematic diagram showing the execution of step S312 by taking the kurtosis factor as an example according to one aspect of the present invention is shown;

[0053] Figure 4 A schematic diagram showing the execution of step S313 by taking the kurtosis factor as an example according to one aspect of the present invention is shown;

[0054] Figure 5 The kurtosis RMS value R of the rolling element defect provided by one aspect of the present invention when the defect size is 0.118 mm is shown. Ku The relationship curve diagram with the data interception length;

[0055] Figure 6 The kurtosis RMS value R of the rolling element defect provided by one aspect of the present invention when the defect size is 0.356 mm is shown. Ku The relationship curve diagram with the data interception length;

[0056] Figure 7 The kurtosis RMS value R of the rolling element defect provided by one aspect of the present invention when the defect size is 0.533 mm is shown. Ku The relationship curve diagram with the data interception length;

[0057] Figure 8 The kurtosis RMS value R of the inner ring defect provided by one aspect of the present invention when the defect size is 0.118 mm is shown. Ku The relationship curve diagram with the data interception length;

[0058] Fig. 9 The kurtosis RMS value R of the inner ring defect provided by one aspect of the present invention when the defect size is 0.356 mm is shown. Ku The relationship curve diagram with the data interception length;

[0059] Fig.10 The kurtosis RMS value R of the inner ring defect provided by one aspect of the present invention when the defect size is 0.533 mm is shown. Ku The relationship curve diagram with the data interception length;

[0060] Fig.11 The kurtosis RMS value R of the outer ring defect provided by one aspect of the present invention when the defect size is 0.118 mm is shown. Ku The relationship curve diagram with the data interception length;

[0061] Fig.12 The kurtosis RMS value R of the inner ring defect provided by one aspect of the present invention when the defect size is 0.356 mm is shown. Ku The relationship curve diagram with the data interception length;

[0062] Fig.13 The kurtosis RMS value R of the inner ring defect provided by one aspect of the present invention when the defect size is 0.533 mm is shown. Ku The relationship curve between the data interception length. DETAILED DESCRIPTION

[0063] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. Note that the aspects described below in conjunction with the accompanying drawings and specific embodiments are only exemplary and should not be construed as limiting the scope of protection of the present invention in any way.

[0064] In the description of the present invention, it should be noted that if the terms "upper", "lower", "inside", "outside", "vertical" and the like appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, or is the orientation or position relationship in which the product of the invention is usually placed when used, rather than indicating or implying that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0065] At the same time, it should be noted that the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0066] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified or limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, an integral connection, or a detachable connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or the internal communication of two components, etc. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0067] Figure 1 This is a flow chart of the bearing condition monitoring parameter screening method provided in this embodiment. Figure 1 An embodiment of the present invention provides a bearing state monitoring parameter screening method, through which statistical characteristic parameters that eliminate data truncation length interference can be screened out as bearing state monitoring parameters for bearing vibration characteristic characterization. The bearing state is characterized by using the bearing state monitoring parameters, and the characterization result is more accurate, which is helpful to identify early defects of the bearing.

[0068] The bearing condition monitoring parameter screening method includes the following steps:

[0069] S01: Determine dimensionless statistical parameter targets for bearing condition monitoring.

[0070] The dimensionless statistical value is not affected by the bearing size parameters, can directly reflect the impact characteristics of bearing defect excitation, has strong universality for bearings of different types and sizes, and some statistical values ​​are sensitive to vibration shock. Therefore, the dimensionless statistical parameters are used as the screening target for early bearing defect identification. Specifically, the statistical characteristic parameters commonly used for state monitoring of the target bearing can be selected as the dimensionless statistical parameter target based on experience, and the dimensionless statistical parameter target includes multiple statistical characteristic parameters.

[0071] Optionally, the plurality of statistical characteristic parameters include a kurtosis factor K u , volatility factor C r , Pulse Factor I m , marginal factor P r and the form factor W a In an embodiment, the plurality of statistical characteristic parameters include a kurtosis factor K u , volatility factor C r , Pulse Factor I m , marginal factor P r and the form factor W a It is understandable that in some other embodiments, the specific types of multiple statistical characteristic parameters included in the dimensionless statistical parameter target can also be set according to requirements, for example, only the kurtosis factor K u , volatility factor Cr , Pulse Factor I m , marginal factor P r and the form factor W a The part in, or can also be set to exclude the kurtosis factor K u , volatility factor C r , Pulse Factor I m , marginal factor P r and the form factor W a In addition, other statistical characteristic parameters are also included.

[0072] S02: Get the data set.

[0073] The historical bearing vibration signal data required for subsequent screening operations constitute the data set, that is, the data set is a collection of historical bearing vibration signal data, which can be bearing data in a public database, such as bearing data in the bearing database of Western Reserve University; or bearing data obtained according to own needs, such as data collected by internal monitoring of the company. It is understandable that the existing bearing database itself can also be used as the data set.

[0074] It should be noted that the target bearing is the bearing that needs to be monitored for bearing condition, and the historical bearing vibration signal data included in the data set is the vibration signal data of the historical bearing of the same type as the target bearing and applied in the same scenario. At the same time, the speed of the historical bearing vibration signal data used as screening data is stable, and the speed stability can be determined based on the conventional judgment standards in this field.

[0075] S03: Obtain bearing condition monitoring parameters for bearing vibration feature characterization in dimensionless statistical parameter targets according to the screening data.

[0076] The screening data is the historical bearing vibration signal data selected from the data set for screening. The historical bearing vibration signal data that meets different conditions can be selected as the screening data according to the screening requirements.

[0077] The screening process includes: segmenting the screened data according to different data lengths, and calculating the time-domain statistical root mean square values ​​corresponding to multiple statistical characteristic parameters based on the segmented screened data; if the time-domain statistical root mean square value meets the first preset condition, it is considered that the statistical characteristic parameter corresponding to the time-domain statistical parameter has passed the screening, that is, the statistical characteristic parameter can be used as a bearing condition monitoring parameter.

[0078] Specifically, the data length is divided according to the number of data points contained therein. For example, if the data length is 1, then it contains 1 data point, and if the data length is 100, then the segment of data contains 100 data points. It should be noted that in bearing vibration monitoring, generally, a data point is obtained at a preset time interval, so the number of data points can be used to characterize the time length corresponding to each segment of data.

[0079] Optionally, the first preset condition is that the time domain statistical root mean square values ​​calculated under different data clipping lengths are stable, or there is a data length clipping range in which the time domain statistical values ​​are stable.

[0080] Specifically, if the data interception length is changed, the calculated time domain statistical RMS value first changes greatly and then tends to be stable after the data interception length reaches a certain value. It is believed that the statistical characteristic parameter corresponding to the time domain statistical RMS value has good stability within the data length interception range and can pass the screening.

[0081] Furthermore, the above-mentioned time-domain statistical root mean square value is stable means that: when the data interception length increases by 100, the change of the time-domain statistical root mean square value does not exceed 10%.

[0082] Optionally, the step of obtaining the bearing condition monitoring parameters for bearing vibration feature characterization in the dimensionless statistical parameter target according to the screening data includes multi-level screening; each level of screening selects historical bearing vibration signal data with preset characteristics from the data set as screening data, and performs the screening process. Finally, the statistical characteristic parameters that pass the screening in the last level of screening are used as bearing condition monitoring parameters. Optionally, the multi-level screening includes the following steps:

[0083] S31: First level screening.

[0084] In the first level screening, the preset feature is any defect. In other words, the screening data in the screening only needs to be confirmed as the vibration data of historical bearings with defects, without limiting the defect type (such as a single defect or a mixed defect), the degree of the defect, etc.

[0085] According to the screening process, multiple statistical characteristic parameters included in the dimensionless statistical parameter target are screened, and the statistical characteristic parameters that pass the screening are used as the first-level screening results. The screening process is described in detail below by taking the first-level screening as an example.

[0086] S311: Divide the screening data into m segments of equal length by length L.

[0087] There are n data points included in the data segment of length L. It should be noted that, in this embodiment, the data length is represented by the number of data points. In other words, in this embodiment, L can be regarded as equal to n.

[0088] S312: Calculate the time domain statistical parameter value corresponding to the statistical characteristic parameter according to the n data points in each data segment to obtain m time domain statistical parameter values.

[0089] The time domain statistical parameter values ​​corresponding to multiple statistical parameter features are calculated according to the following formulas:

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] Where j represents the jth data segment, j = 1 to m; x(t) is the number of n discrete data points {x1, x2, x3…x n} is a one-dimensional array.

[0100] Since the screened data is divided into m data segments, m time-domain statistical parameter values ​​can be obtained by performing the above calculation on each data segment respectively.

[0101] The time domain statistical parameter values ​​corresponding to the m data segments can be expressed as:

[0102]

[0103] S313: Calculate a root mean square value according to the m time-domain statistical parameter values ​​to obtain a time-domain statistical root mean square value.

[0104] Specifically, by {K u1 ,K u2 ,K u3 …K uj …K um} to calculate the RMS value and obtain the time domain statistical RMS value corresponding to the kurtosis factor; by {C r1 ,C r2 ,C r3 …C rj …C rm} to calculate the RMS value and obtain the time domain statistical RMS value corresponding to the fluctuation factor; m1 ,I m2 ,I m3 …I mj …I mm} to calculate the RMS value and obtain the time domain statistical RMS value corresponding to the pulse factor; r1 ,P r2 ,P r3 …P rj …P rm} to calculate the RMS value and obtain the time domain statistical RMS value corresponding to the marginal factor; a1 ,W a2 ,W a3 …W aj …W am}Calculate the RMS value to obtain the time domain statistical RMS value corresponding to the waveform factor.

[0105] During the calculation process, the m time-domain statistical parameter values ​​can be divided sequentially. For example, when m is 1000, every 100 time-domain statistical values ​​are divided into one section in order, and the root mean square value calculated from the time-domain statistical parameter values ​​contained in one section is used as the time-domain statistical root mean square value. In other words, 10 time-domain statistical root mean square values ​​can be obtained at this time, and the 10 time-domain statistical root mean square values ​​change periodically.

[0106] S314: Repeat steps S311 to S313 for multiple times, and change the data length in step S311, so as to obtain the time domain statistical root mean square value under different data lengths.

[0107] S315: Compare the time-domain statistical RMS values ​​obtained in step S314. If the time-domain statistical RMS values ​​are stable, or the time-domain statistical RMS values ​​are stable within a certain interception length range (i.e., they meet the first preset condition), then the statistical characteristic parameter corresponding to the time-domain statistical RMS value passes the screening. On the contrary, if the time-domain statistical parameter RMS value is consistently unstable during the change of the data interception length (i.e., it does not meet the first preset condition), it means that the statistical parameter is not suitable for the characterization of defect characteristics and can be directly excluded if it fails the screening.

[0108] S32: Second level screening.

[0109] In the second level screening, the preset feature is a preset defect type. The specific type of the preset defect type can be selected according to the needs, for example, the preset defect type can be set to an inner ring defect, an outer ring defect or a rolling element defect. The historical bearing vibration signal data with the preset defect type is selected from the data set as the screening data, and the screening process is performed.

[0110] When performing the screening process, the first-level screening results are screened. In other words, in the second-level screening, only the corresponding time-domain statistical root mean square values ​​of the statistical characteristic parameters contained in the first-level screening results need to be calculated. The statistical characteristic parameters that pass the second-level screening are used as the second-level screening results. At the same time, the data interception length range of the statistical characteristic parameters whose time-domain statistical root mean square values ​​are stable only within a certain data interception length range during the screening process can be obtained.

[0111] Through the second level screening, the influence of defect type on the data interception length range is eliminated.

[0112] S33: Third level screening.

[0113] In the third level screening, the preset features are preset defect types and preset defect levels, and the preset defect levels are set to be slight, medium or severe according to the defect size. For example, if the preset defect type set in step S32 is an inner ring defect, and the preset defect level set in step S33 is slight, then historical bearing vibration signal data with inner ring defects and slight defects are selected from the data set as screening data, and the screening process is performed.

[0114] When performing the screening process, screening is performed from the secondary screening results. In other words, in the third-level screening, only the corresponding time-domain statistical root mean square value needs to be calculated for the statistical characteristic parameters contained in the secondary screening results. The statistical characteristic parameters that pass the third-level screening are used as the third-level screening results. At the same time, the data interception length range of the statistical characteristic parameters whose time-domain statistical root mean square value is stable only within a certain data interception length range during the screening process can be obtained.

[0115] Through the third level screening, the influence of the defect degree on the data interception length range is eliminated.

[0116] S34: Fourth level screening.

[0117] In the fourth level screening, the preset features are preset defect type, preset defect degree and preset speed, and the specific value of the preset speed can be set according to the normal working speed of the target bearing, etc. For example, the preset defect type set in step S32 is inner ring defect, the preset defect degree set in step S33 is slight, and the preset speed set in step S34 is 1730r / min, then the historical bearing vibration signal data with inner ring defect, slight defect and speed of 1730r / min is selected from the data set as screening data, and the screening process is performed.

[0118] In the screening process, screening is performed from the three-level screening results. In other words, in the fourth-level screening, only the corresponding time-domain statistical root mean square value of the statistical characteristic parameters contained in the three-level screening results needs to be calculated. The statistical characteristic parameters that pass the fourth-level screening are used as the fourth-level screening results. At the same time, the data interception length range of the statistical characteristic parameters whose time-domain statistical root mean square value is stable only within a certain data interception length range during the screening process can be obtained.

[0119] Through the four-level screening, the influence of rotation speed on the data interception length range can be eliminated.

[0120] It should be noted that the steps S32 to S34 may be repeatedly performed multiple times after changing the preset features to eliminate the influence of data capture lengths at different defect types, different defect degrees and different rotation speeds on the time domain statistical RMS value.

[0121] In this embodiment, step S03 includes primary screening and secondary screening. The specific steps of the primary screening are the above-mentioned multi-level screening according to the screening process, that is, the contents of the above-mentioned steps S31 to S34. The last level screening result in the multi-level screening is used as the primary screening result. The secondary screening is performed on the basis of the primary screening, and the bearing state monitoring parameters are obtained. It should be noted that in this embodiment, the final bearing state monitoring parameters are obtained through primary screening and secondary screening. It can be understood that in other embodiments, if only one screening is included, the last level screening result in the primary screening can also be used as the bearing state monitoring parameters.

[0122] Optionally, the specific steps of the secondary screening include:

[0123] A plurality of historical bearing vibration signal data with preset defect types and different defect sizes are selected from the data set as screening data, and it is determined whether the relationship between the defect size corresponding to the screening data and the time domain statistical root mean square value corresponding to the statistical characteristic parameter in the first screening result meets the second preset condition. If the second preset condition is met, the statistical characteristic parameter corresponding to the time domain statistical root mean square value can be used as a bearing condition monitoring parameter to perform bearing condition monitoring.

[0124] Specifically, the data cutoff length range obtained in step S33 is used to segment the multiple screening data in the secondary screening, and the amplitude of the time domain statistical root mean square value calculated under different defect sizes is obtained. If within a certain defect size range, the defect size is positively correlated with the amplitude of the time domain statistical root mean square value, that is, when the defect size increases, the amplitude of the time domain statistical root mean square value also increases, then it means that the time domain statistical root mean square value meets the second preset condition and can more accurately reflect the degree of bearing defects. The corresponding statistical characteristic parameters can be used as bearing condition monitoring parameters.

[0125] The following is an example of the calculation of the time domain statistical root mean square value corresponding to the kurtosis factor to illustrate the above calculation process:

[0126] Figure 2 FIG. 4 is a schematic diagram showing the execution of step S311 by taking the kurtosis factor as an example. Figure 2 In the example, the horizontal axis is time and the vertical axis is vibration signal. Figure 2 The vibration signal data shown is divided into 3 segments according to the data interception length L, and each segment contains 7 data points. Figure 3 FIG. 4 is a schematic diagram showing the execution of step S312 by taking the kurtosis factor as an example. Figure 3 In the figure, the horizontal axis is time and the vertical axis is the kurtosis value K. u (i.e., the time domain statistical parameter value corresponding to the kurtosis factor), Figure 2 Calculate the kurtosis value K for the three segments of data u Get K u1 , K u2 and K u3 . Figure 4 A schematic diagram showing the execution of step S313 using the kurtosis factor as an example is shown. Figure 3 In the figure, the horizontal axis is time and the vertical axis is the RMS value of kurtosis R Ku (That is, the time domain statistical root mean square value corresponding to the kurtosis factor).

[0127] Figure 5 The RMS value R of the rolling element defect when the defect size is 0.118 mm is shown. Ku The relationship curve between the length of the data interception, Figure 6 The RMS value R of the rolling element defect when the defect size is 0.356 mm is shown. Ku The relationship curve between the length of the data interception, Figure 7 The RMS value R of the rolling element defect when the defect size is 0.533 mm is shown. Ku The relationship curve between the length of the data interception, Figure 8 The RMS value R of the inner ring defect when the defect size is 0.118 mm is shown. Ku The relationship curve between the length of the data interception, Fig. 9 The RMS value R of the inner ring defect when the defect size is 0.356 mm is shown. Ku The relationship curve between the length of the data interception, Fig.10 The RMS value R of the inner ring defect when the defect size is 0.533 mm is shown. Ku The relationship curve between the length of the data interception, Fig.11 The RMS value R of the outer ring defect when the defect size is 0.118 mm is shown. KuThe relationship curve between the length of the data interception, Fig.12 The RMS value R of the inner ring defect when the defect size is 0.356 mm is shown. Ku The relationship curve between the length of the data interception, Fig.13 The RMS value R of the inner ring defect when the defect size is 0.533 mm is shown. Ku The relationship curve between the length of the data interception. Figure 5-Figure 13 It can be seen that the time domain statistical RMS value corresponding to the kurtosis factor is only stable within a certain data interception length range, and the data interception degree has an impact on the characteristic representation of the kurtosis factor. At the same time, under different defect types, different defect sizes and different rotation speeds, the data interception length range corresponding to the stability of the time domain statistical RMS value is different.

[0128] In the bearing state monitoring parameter screening method provided by the embodiment of the present invention, the dimensionless statistical characteristic parameters can intuitively reflect the impact characteristics, and are more suitable for the characterization of the instantaneous impact characteristics of the bearing. At the same time, the corresponding time-domain statistical root mean square value can be defined to describe the overall energy represented by the statistical characteristic parameters, reflecting the vibration impact amplitude level of the defective bearing within the process time, and used to estimate the degree of bearing defects. Due to the amplification of the vibration characteristics by the statistical characteristic parameters, compared with directly observing the changes in the bearing vibration amplitude, the time-domain statistical root mean square value of the statistical characteristic parameters proposed in this article has a greater advantage in reflecting the degree of bearing defects. Moreover, the bearing state monitoring parameter screening method fully considers the coupled excitation effects of defect types and defect degrees. Therefore, the bearing state monitoring parameter screening method provided by the present invention can screen and obtain more accurate and reliable monitoring results.

[0129] An embodiment of the present invention further provides a bearing condition monitoring method, which includes the following steps:

[0130] Obtain bearing state monitoring parameters for bearing vibration feature characterization. The bearing state monitoring parameters can be obtained through the above-mentioned bearing state monitoring parameter screening method.

[0131] The bearing state is monitored according to the bearing state monitoring parameters. That is, in the bearing state monitoring process, the bearing vibration characteristics are characterized by using the above-mentioned bearing state monitoring parameters.

[0132] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technology in the field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for screening bearing condition monitoring parameters, characterized in that: The bearing condition monitoring parameter screening method comprises: Determining a dimensionless statistical parameter target for bearing condition monitoring; wherein the dimensionless statistical parameter target includes a plurality of statistical characteristic parameters; Acquire a data set; wherein the data set includes historical bearing vibration signal data; According to the screening data, a bearing condition monitoring parameter for bearing vibration feature characterization is obtained in the dimensionless statistical parameter target; wherein the screening data is historical bearing vibration signal data selected from the data set for screening; the screening process includes: Segmenting the screening data according to different data lengths, and calculating the time-domain statistical root mean square values ​​corresponding to the multiple statistical characteristic parameters according to the segmented screening data; If the time-domain statistical RMS value satisfies the first preset condition, the statistical characteristic parameter corresponding to the time-domain statistical RMS value passes the screening.

2. The bearing condition monitoring parameter screening method according to claim 1, characterized in that: The step of obtaining bearing condition monitoring parameters that can be used to characterize bearing vibration characteristics in the dimensionless statistical parameter target based on the screening data includes multi-level screening; each level of screening selects the historical bearing vibration signal data with preset characteristics from the data set as the screening data and executes the screening process; and uses the statistical characteristic parameters that pass the screening in the last level of screening as the bearing condition monitoring parameters.

3. The bearing condition monitoring parameter screening method according to claim 2 is characterized in that: The multi-level screening includes a first-level screening; the preset feature of the first-level screening is any defect; the first-level screening includes: The plurality of statistical characteristic parameters are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as primary screening results.

4. The bearing condition monitoring parameter screening method according to claim 3 is characterized in that: The multi-level screening also includes a second-level screening; the preset feature of the second-level screening is having a preset defect type; The second level of screening includes: The first-level screening results are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as the second-level screening results.

5. The method for screening bearing condition monitoring parameters according to claim 4, characterized in that: The multi-level screening also includes a third-level screening; the preset characteristics of the third-level screening are having the preset defect type and having a preset defect degree; The third level screening includes: The secondary screening results are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as the tertiary screening results.

6. The method for screening bearing condition monitoring parameters according to claim 5, characterized in that: The multi-level screening also includes a fourth-level screening; the preset characteristics of the fourth-level screening are having the preset defect type and the preset defect degree and having a preset rotation speed; the fourth-level screening includes: The third-level screening results are screened according to the screening process, and the statistical characteristic parameters that pass the screening are used as the fourth-level screening results.

7. The method for screening bearing condition monitoring parameters according to claim 1, characterized in that: The first preset condition includes: the time domain statistical root mean square value calculated under different data interception lengths is stable, or there is a data length interception range in which the time domain statistical root mean square value is stable.

8. The method for screening bearing condition monitoring parameters according to claim 7, characterized in that: The time domain statistical root mean square value is stable means that when the data interception length increases by 100, the change of the time domain statistical root mean square value does not exceed 10%.

9. The method for screening bearing condition monitoring parameters according to claim 1, characterized in that: The steps of segmenting the screened data and calculating the time domain statistical root mean square values ​​corresponding to the multiple statistical characteristic parameters according to each segment of the data include: The screening data is divided into m segments of equal length by length L; each data segment of length L contains n data points; Calculating the time domain statistical parameter value corresponding to the statistical characteristic parameter according to the n data points in each data segment to obtain m time domain statistical parameter values; A root mean square value is calculated according to the m time-domain statistical parameter values ​​to obtain the time-domain statistical root mean square value.

10. The method for screening bearing condition monitoring parameters according to claim 9, characterized in that: The plurality of statistical characteristic parameters include a kurtosis factor K u , volatility factor C r , Pulse Factor I m , marginal factor P r and the form factor W a at least part of .

11. The method for screening bearing condition monitoring parameters according to claim 10, characterized in that: The formula for calculating the time domain statistical parameter value corresponding to the statistical characteristic parameter according to n data points in each data segment is as follows: Where j represents the jth data segment; x(t) is the number of n discrete data points {x1, x2, x3…x n } constitutes a one-dimensional array; The time domain statistical parameter values ​​corresponding to the m data segments can be expressed as:

12. The method for screening bearing condition monitoring parameters according to claim 1, characterized in that: The step of obtaining the bearing condition monitoring parameters for bearing vibration feature characterization in the dimensionless statistical parameter target according to the screening data comprises: Primary screening: the primary screening is performed through the screening process to obtain a primary screening result; Secondary screening: select historical bearing vibration signal data with a preset defect size as screening data; determine whether the relationship between the preset defect size corresponding to the screening data and the time domain statistical root mean square value corresponding to the statistical characteristic parameter in the first screening result meets the second preset condition; if the second preset condition is met, use the statistical characteristic parameter corresponding to the time domain statistical root mean square value as the bearing state monitoring parameter.

13. A bearing condition monitoring method, characterized in that: The bearing condition monitoring method comprises: Acquire bearing state monitoring parameters for bearing vibration characteristic characterization; wherein the bearing state monitoring parameters are obtained according to the bearing state monitoring parameter screening method according to any one of claims 1 to 12; The bearing state is monitored according to the bearing state monitoring parameters.

Images