CN117909692A - Intelligent analysis method for operation data of double-disc motor bearing - Google Patents
Intelligent analysis method for operation data of double-disc motor bearing Download PDFInfo
- Publication number
- CN117909692A CN117909692A CN202410303752.9A CN202410303752A CN117909692A CN 117909692 A CN117909692 A CN 117909692A CN 202410303752 A CN202410303752 A CN 202410303752A CN 117909692 A CN117909692 A CN 117909692A
- Authority
- CN
- China
- Prior art keywords
- temperature data
- data
- temperature
- double
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 claims abstract description 46
- 230000002159 abnormal effect Effects 0.000 claims abstract description 42
- 239000013598 vector Substances 0.000 claims abstract description 9
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000000052 comparative effect Effects 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 19
- 230000005856 abnormality Effects 0.000 description 19
- 238000004364 calculation method Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 239000004519 grease Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to the technical field of data processing, and provides an intelligent analysis method for operation data of a double-disc motor bearing, which comprises the following steps: acquiring temperature data and corresponding rotating speed data; acquiring abnormal possibility of the temperature data according to the temperature data; establishing a first comparison straight line according to the temperature data and the corresponding rotating speed data, acquiring a fitting value of the temperature data, establishing a vector corresponding to the corresponding temperature data and the corresponding rotating speed data, determining a deviation included angle, and further acquiring the credibility of the temperature data; and determining the weight of the local density in the LOF value acquisition process according to the credibility and the abnormal possibility of the temperature data, acquiring the LOF value of each temperature data according to the weight of the local density, and completing intelligent analysis of the running data of the double-disk motor bearing according to the LOF value of the temperature data. The invention aims to solve the problems of inaccurate and unstable detection of the potential problem of the double-disc motor bearing operation.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an intelligent analysis method for operation data of a double-disc motor bearing.
Background
The double-disc motor bearing is a bearing for high-speed rotating equipment such as a motor, a generator and the like, and generally consists of two disc bearings, namely an inner disc and an outer disc. The double-disc motor bearing can reduce the influence of axial and radial loads on the inner disc, and improve the bearing capacity and running stability of the bearing. The bearing is an important component of the operation of the double-disc motor, and when the motor bearing fails, the double-disc motor bearing can suddenly stop or lose control, and workers are easy to fall down, be clamped or be damaged. Therefore, the potential problems of the double-disc motor bearing need to be found and solved in time, and the normal operation of the double-disc motor bearing and the safety of a workplace are ensured.
In order to monitor the potential problem of the double-disc motor bearing, the temperature data of the disc motor bearing is generally subjected to anomaly detection by using an LOF anomaly detection algorithm, but the temperature data of the disc motor bearing has small difference, and when the difference between the temperature data in the normal temperature change range and other temperature data is relatively large, the temperature data in the normal temperature change range is easy to be misjudged as anomaly data. Therefore, a more stable method for detecting the abnormality of the temperature data of the double-disc motor bearing is needed, so that the potential problem detection of the double-disc motor bearing is more accurate and stable.
Disclosure of Invention
The invention provides an intelligent analysis method for operation data of a double-disc motor bearing, which aims to solve the problems of inaccurate and unstable detection of potential problems of the operation of the double-disc motor bearing caused by the fact that the whole operation data of the existing double-disc motor bearing is relatively close, and adopts the following technical scheme:
The embodiment of the invention provides an intelligent analysis method for double-disc motor bearing operation data, which comprises the following steps:
Acquiring temperature data of a motor, preprocessing the temperature data, and acquiring rotating speed data corresponding to the temperature data;
establishing a temperature data set according to the temperature data, acquiring the neighbor radius of the temperature data, determining the neighbor temperature value and the local reachable density of the temperature data, and acquiring the abnormal possibility of the temperature data according to the neighbor radius, the neighbor temperature value and the local reachable density of the temperature data in the temperature data set;
Establishing a temperature-rotating speed plane rectangular coordinate system, acquiring a linear relation corresponding to temperature data and rotating speed data, acquiring a fitting value of each temperature data, establishing a first comparison straight line and vectors corresponding to the corresponding temperature data and rotating speed data, determining a deviation included angle of the corresponding temperature data and rotating speed data, and acquiring the credibility of the temperature data according to the abnormal possibility of the temperature data contained in the adjacent radius of the temperature data, the deviation included angle of the temperature data and the rotating speed data contained in the adjacent radius of the temperature data and the fitting value of the temperature data;
And determining the weight of the local density in the LOF value acquisition process according to the credibility and the abnormal possibility of the temperature data, acquiring the LOF value of each temperature data according to the weight of the local density, and completing intelligent analysis of the running data of the double-disk motor bearing according to the LOF value of the temperature data.
Further, the method for establishing the temperature data set according to the temperature data comprises the following specific steps: the set of temperature data at all acquisition times is denoted as a temperature dataset.
Further, the method for acquiring the neighbor radius of the temperature data comprises the following specific steps:
And respectively marking each temperature data as temperature data to be analyzed, marking first preset threshold value data closest to the numerical value of the temperature data to be analyzed as neighbor edge data of the temperature data to be analyzed, and marking the absolute value of the maximum value of the difference value between the neighbor edge data of the temperature data to be analyzed and the temperature data to be analyzed as the neighbor radius of the temperature data to be analyzed.
Further, the determining the neighbor temperature value and the local reachable density of the temperature data comprises the following specific methods:
and recording the average value of all the temperature data contained in the neighbor radius of the temperature data to be analyzed as the neighbor temperature value of the temperature data to be analyzed.
Further, the method for establishing the rectangular coordinate system of the temperature-rotating speed plane comprises the following specific steps: and establishing a temperature-rotating speed plane rectangular coordinate system by taking the temperature data as a vertical axis and the rotating speed data as a horizontal axis.
Further, the method for obtaining the linear relation corresponding to the temperature data and the rotation speed data comprises the following specific steps: and performing linear fitting by taking the temperature data as a dependent variable and the rotating speed data as an independent variable to obtain a linear relation corresponding to the temperature data and the rotating speed data.
Further, the method for obtaining the first comparison line comprises the following steps: in a rectangular coordinate system of a temperature-rotating speed plane, a straight line corresponding to a linear relation between temperature data and rotating speed data is recorded as a first comparison straight line.
Further, the specific method for determining the deviation included angle of the corresponding temperature data and the corresponding rotation speed data comprises the following steps:
And recording the included angle between the vector corresponding to the corresponding temperature data and the rotating speed data and the first comparison straight line as the deviated included angle of the group of corresponding temperature data and rotating speed data.
Further, the determining the weight of the local density in the LOF value obtaining process according to the credibility and the possibility of abnormality of the temperature data, and obtaining the LOF value of each temperature data according to the weight of the local density comprises the following specific steps:
In the method, in the process of the invention, Represents the/>Temperature data/>In the LOF value acquisition process, the weight of the local density; /(I)Represents the/>Temperature data/>Is the degree of confidence of (2); /(I)Represents the/>Temperature data/>Is an abnormal possibility of (1); /(I)Representing a first adjustment parameter; /(I)The representation contains the/>Neighborhood number of individual temperature data; /(I)The representation contains the/>First/>, of the temperature dataThe credibility of the temperature data corresponding to each neighborhood; /(I)The representation contains the/>First/>, of the temperature dataLocal reachable density of temperature data corresponding to each neighborhood;
In the method, in the process of the invention, Represents the/>Temperature data/>The LOF value of (2); /(I)Representation/>Is a local reachable density of (3); /(I)Representing a first preset threshold; /(I)Represents the/>Temperature data/>Included in the neighborhood of (1)A weight of the local density of the individual temperature data; represents the/> Temperature data/>Included in the neighborhood of (1)And temperature data.
Further, the intelligent analysis of the operation data of the double-disc motor bearing is completed according to the LOF value of the temperature data, and the method comprises the following specific steps:
when the LOF value of the temperature data is larger than or equal to the selection threshold value, the temperature data is considered to be abnormal data;
when abnormal data are detected, the double-disc motor bearing is considered to have potential problems, the double-disc motor bearing needs to be overhauled in time, and otherwise, the double-disc motor bearing is considered to operate normally.
The beneficial effects of the invention are as follows:
According to the invention, corresponding temperature data and rotation speed data acquired at the same acquisition time are analyzed, firstly, because the double-disc motor bearing often generates abnormal heat when the double-disc motor bearing works under potential risk, the reliability of the temperature data is evaluated according to the possibility that the temperature data shows abnormal temperature data, namely the abnormal possibility of acquiring the temperature data; then, according to the characteristic that the temperature of the bearing is increased and the temperature data is influenced due to mutual friction between the bearing and the journal, so that the abnormal possibility of the larger temperature data caused by normal operation of the double-disk motor bearing is overlarge, the credibility of the temperature data, namely the credibility of the temperature data reflecting the actual working condition of the double-disk motor bearing and the possibility that the temperature data is the temperature data acquired when the double-disk motor bearing normally operates, is determined according to the corresponding temperature data and the corresponding rotating speed data, and the accuracy and the stability of the judgment of the operation result of the double-disk motor bearing can be improved; finally, because a neighborhood is required to be constructed in the density calculation process of data in the LOF anomaly detection algorithm, the same data can be contained in a plurality of neighbors, each data can participate in the calculation of the reachable densities of a plurality of data, by utilizing the characteristic, the weight of the local density in the LOF value acquisition process is determined according to the credibility and the anomaly possibility of the temperature data, the greater weight is given to the temperature data which is more likely to be acquired under normal conditions, the difference between the weights corresponding to different temperature data is increased, the LOF value of the temperature data with the greater difference is further acquired, the intelligent analysis of the operation data of the double-disc motor bearing is completed according to the LOF value of the temperature data, and the problems of inaccurate and unstable detection of the potential problem of the operation of the double-disc motor bearing caused by the overall approaching of the operation data of the double-disc motor bearing are solved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a flow chart of a method for intelligently analyzing operation data of a double-disk motor bearing according to an embodiment of the present invention;
Fig. 2 is an abnormality possibility acquisition flowchart.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a method for intelligently analyzing operation data of a dual-disc motor bearing according to an embodiment of the invention is shown, the method includes the following steps:
And S001, acquiring temperature data of the motor, preprocessing the temperature data, acquiring the temperature data, and acquiring rotating speed data corresponding to the temperature data.
Temperature is a key parameter indicating the state and performance of a double-disk motor bearing, and temperature data is one of early indexes capable of reflecting motor bearing faults or grease lubrication problems. Meanwhile, certain correlation exists between the temperature of the bearing and other indexes, wherein the other indexes comprise vibration, noise, rotating speed and the like, and when the temperature is abnormal, the temperature data often influence the performance of the other indexes. Therefore, the monitoring of potential safety hazards possibly existing in the motor can be achieved through the abnormality of the monitoring temperature data.
And taking half an hour as an empirical value of a time interval, acquiring motor temperature data by using a temperature sensor every 1 minute, and recording the average value of all the motor temperature data in the time interval as the temperature data of the next acquisition time in two acquisition times corresponding to the time interval.
The running temperature difference of the double-disc motor bearing in normal operation is not large, so that the temperature data acquired from the double-disc motor bearing in normal operation is relatively concentrated, the temperature data can be abnormally detected by using an LOF abnormality detection algorithm, and when abnormal temperature data are detected, the double-disc motor bearing is considered to have potential problems. However, only the operation result of the double-disc motor bearing is directly obtained according to the temperature data, and the micro-variation of the temperature data often causes great difference in the operation result of the double-disc motor bearing, so that the rotation speed data at each acquisition moment is obtained, the rotation speed data is analyzed while the temperature data is analyzed, and the accuracy and the stability of the judgment of the operation result of the double-disc motor bearing are improved.
And acquiring the rotation speed data of each acquisition time by using a rotation speed sensor, wherein each acquisition time corresponds to one temperature data and one rotation speed data.
So far, corresponding temperature data and rotating speed data are obtained.
Step S002, a temperature data set is established according to the temperature data, the neighbor radius of the temperature data is obtained, the neighbor temperature value and the local reachable density of the temperature data are determined, and the abnormal possibility of the temperature data is obtained according to the neighbor radius, the neighbor temperature value and the local reachable density of the temperature data in the temperature data set.
In order to improve accuracy and stability of operation result judgment of the double-disc motor bearing, temperature data are analyzed, meanwhile, rotating speed data are analyzed, reliability of different temperature data is judged, weight of the different temperature data in the abnormality detection process is determined according to the reliability, namely, the LOF value in the LOF abnormality detection algorithm is determined by weighting the reachable density in the LOF abnormality detection algorithm. The LOF anomaly detection algorithm is a known technique and will not be described in detail.
First, the temperature data is analyzed for the possibility of exhibiting abnormal temperature data, i.e., the credibility of different temperature data.
The temperature data collected in the normal working process of the double-disc motor bearing generally fluctuates in a smaller range, when the double-disc motor bearing has potential risks, the collected temperature data deviate more from the normal fluctuation range, and the fluctuation degree of all the temperature data can be enlarged by the collected temperature data at the moment, so that the local density of the collected temperature data in the normal working process changes.
In this embodiment, k of the k nearest neighbor distance in the LOF anomaly detection algorithm is taken as a first preset threshold, and the empirical value of the first preset threshold is 9.
The set of temperature data at all acquisition times is denoted as a temperature dataset.
Each of the temperature data sets is analyzed separately.
The first preset threshold value of the temperature data closest to the value of the temperature data is recorded as the adjacent edge data of the analyzed temperature data, and the absolute value of the maximum value of the difference value between the adjacent edge data of the temperature data and the temperature data is recorded as the adjacent radius of the analyzed temperature data.
The average of all the temperature data contained within the neighbor radius of the analyzed temperature data is noted as the neighbor temperature value of the analyzed temperature data.
A local reachable density of each of the temperature data sets is obtained using a LOF anomaly detection algorithm.
The smaller the local reachable density of the temperature data, the greater the likelihood that the temperature data will be obtained during operation of the double disk motor bearing at potential risk.
And acquiring abnormal possibility of the temperature data according to the neighbor radius, the neighbor temperature value and the local reachable density of the temperature data in the temperature data set.
Wherein,Represents the temperature dataset/>Temperature data/>Is an abnormal possibility of (1); /(I)Representing the temperature datasetA plurality of temperature data; /(I)Representation/>Is a neighbor radius of (2); /(I)Representation/>Is a local reachable density of (3); /(I)Representing a first adjustment parameter, the empirical value being 1; /(I)Representing a first preset threshold value, and taking an empirical value of 9; /(I)Representation/>Included within the neighbor radius of (c)A plurality of temperature data; /(I)Representation/>Is a neighbor temperature value of (a).
When the double-disc motor bearing runs stably, the difference between the collected temperature data is small, so that the possibility that the temperature data is abnormal is reflected by the difference between the temperature data. When the difference between the temperature data included in the neighboring radii of the temperature data in the temperature data set is larger, the temperature data corresponds toThe greater the likelihood of anomalies in the temperature data, the greater the likelihood that the temperature data will be obtained during operation of the double-disk motor bearing at potential risk.
When the double-disc motor bearing works under the potential risk, the double-disc motor bearing often generates abnormal heat, so when the adjacent temperature value of the temperature data is larger and the local reachable density is smaller, the abnormal possibility of the temperature data is larger, and at the moment, the possibility that the temperature data is acquired in the working process of the double-disc motor bearing under the potential risk is larger.
To this end, the possibility of abnormality of all the temperature data in the temperature data set is acquired, and the possibility of abnormality acquisition flowchart is shown in fig. 2.
Step S003, a temperature-rotating speed plane rectangular coordinate system is established, a linear relation corresponding to temperature data and rotating speed data is obtained, a fitting value of each temperature data is obtained, a first comparison straight line and vectors corresponding to the corresponding temperature data and rotating speed data are established, a deviation included angle of the corresponding temperature data and rotating speed data is determined, and the credibility of the temperature data is obtained according to the abnormal possibility of the temperature data contained in the adjacent radius of the temperature data, the deviation included angle of the temperature data and the rotating speed data contained in the adjacent radius of the temperature data and the fitting value of the temperature data.
When the double-disc motor bearing rotates, mutual friction between the bearing and the shaft neck can lead to temperature rise of the bearing, meanwhile, the rotation of the bearing can lead to degradation and loss of motor grease to a certain extent, and further heat generated by friction among all components of the friction bearing is more, so that the temperature rises and temperature data are influenced. Therefore, the abnormal data information of the bearing can not be accurately reflected only through single temperature data change, and the temperature rise can be the influence caused by the normal operation of the double-disc motor bearing. Therefore, not every temperature data with a large possibility of abnormality is abnormal data, and the large possibility of abnormality may also be caused by normal operation of the double-disk motor bearing.
In order to eliminate the fact that larger temperature data caused by normal operation of the double-disc motor bearing are misjudged as abnormal data, whether the change of the temperature data is a normal phenomenon caused by normal operation of the double-disc motor bearing is judged according to the relation between the temperature data and the rotating speed data, wherein the relation between the temperature data and the rotating speed data is that the larger the rotating speed of the double-disc motor bearing is, the higher the temperature acquired at the same moment is.
And establishing a temperature-rotating speed plane rectangular coordinate system by taking the temperature data as a vertical axis and the rotating speed data as a horizontal axis.
Fitting all the corresponding temperature data and rotation speed data by using a least square method to obtain a linear relation corresponding to the temperature data and the rotation speed data, wherein the temperature data is a dependent variable, and the rotation speed data is an independent variable. And obtaining a fitting value of each temperature data according to a linear relation corresponding to the temperature data and the rotating speed data.
And drawing a straight line corresponding to a linear relation formula corresponding to the temperature data and the rotating speed data in a temperature-rotating speed plane rectangular coordinate system, and marking the drawn straight line as a first comparison straight line. Marking the point corresponding to each corresponding temperature data and rotation speed data in a rectangular coordinate system of the temperature-rotation speed plane, and marking the vector taking the origin of the coordinates as a starting point and the point corresponding to the corresponding temperature data and rotation speed data as an end point as the vector corresponding to the corresponding temperature data and rotation speed data.
And recording the included angle between the vector corresponding to the corresponding temperature data and the rotating speed data and the first comparison straight line as the deviated included angle of the group of corresponding temperature data and rotating speed data.
And acquiring the credibility of the temperature data according to the abnormal possibility of the temperature data contained in the adjacent radius of the temperature data, the deviation included angle between the temperature data and the rotating speed data contained in the adjacent radius of the temperature data and the fitting value of the temperature data.
Wherein,Represents the temperature dataset/>Temperature data/>Is the degree of confidence of (2); /(I)Representing a first preset threshold value, and taking an empirical value of 9; /(I)Represents the temperature dataset/>First/>, contained within neighbor radii of individual temperature dataAbnormal likelihood of the individual temperature data; /(I)Representing a first adjustment parameter, the empirical value being 1; /(I)Represents the/>Temperature data/>And the corresponding deviation included angle of the rotating speed data; /(I)Representation/>Included within the neighbor radius of (c)Temperature data and/>Deviating included angles of the rotating speed data corresponding to the temperature data; /(I)Representation/>Included within the neighbor radius of (c)Fitting values of the individual temperature data; /(I)Representation/>Included within the neighbor radius of (c)A plurality of temperature data; /(I)The expression is to take the natural constant as the base and the numerical value in brackets as the exponent.
When the probability of abnormality of the temperature data contained in the adjacent radius of the temperature data is smaller, the degree of abnormal interference of the temperature data is smaller, and the temperature data can reflect the real working condition of the double-disc motor bearing.
When the deviation included angles corresponding to the temperature data contained in the adjacent radius of the temperature data are smaller, the correlation between the temperature data contained in the adjacent radius of the temperature data and the corresponding rotating speed data is more consistent with the relationship between the temperature data and the rotating speed data, which are characterized in normal operation of the disc motor bearing, namely, the more likely the temperature data are the temperature data collected in normal operation of the double-disc motor bearing.
When the abnormal possibility of the temperature data contained in the adjacent radius of the temperature data is smaller, the deviation included angle between the temperature data and the rotating speed data contained in the adjacent radius of the temperature data is smaller, and the difference between the fitting value of the temperature data and the temperature data is smaller, the reliability degree of the temperature data is larger, at the moment, the temperature data can reflect the real working condition of the double-disc motor bearing, and is more likely to be the temperature data acquired when the double-disc motor bearing works normally.
So far, the credibility of all the temperature data in the temperature data set is obtained.
Step S004, determining the weight of the local density in the LOF value acquisition process according to the credibility and the abnormal possibility of the temperature data, acquiring the LOF value of each temperature data according to the weight of the local density, and completing the intelligent analysis of the double-disk motor bearing operation data according to the LOF value of the temperature data.
When the double-disc motor bearing has potential risks, the collected temperature data deviate from the normal fluctuation range, at the moment, the collected temperature data can enable the local density of the temperature data in the normal fluctuation range to change, and meanwhile, when abnormal conditions of the temperature data deviating from the normal fluctuation range are different, the influence degree on the local density of the temperature data is different.
In the LOF anomaly detection algorithm, a neighborhood is required to be constructed in the calculation process of the density of the data, so that the same data can be contained in a plurality of neighbors, each data can participate in the calculation of the reachable densities of a plurality of data, and the feature is used for determining the weight of the reachable densities of each temperature data when the LOF value is calculated.
And determining the weight of local density in the LOF value acquisition process in the LOF abnormality detection algorithm according to the abnormality possibility and the reliability degree of the temperature data, and improving the monitoring capability of the LOF abnormality detection algorithm on the abnormal data and the algorithm stability.
In the method, in the process of the invention,Represents the/>Temperature data/>In the LOF value acquisition process, the weight of the local density; /(I)Represents the/>Temperature data/>Is the degree of confidence of (2); /(I)Represents the/>Temperature data/>Is an abnormal possibility of (1); /(I)Representing a first adjustment parameter, the empirical value being 1; /(I)The representation contains the/>Neighborhood number of individual temperature data; /(I)The representation contains the/>First/>, of the temperature dataThe credibility of the temperature data corresponding to each neighborhood; /(I)The representation contains the/>First/>, of the temperature dataLocal reachable densities of temperature data corresponding to the respective neighborhoods.
When the reliability of the temperature data is higher, the temperature data can accurately reflect the real situation, and the higher the weight of the temperature data is given, the higher the weight of the local density corresponding to the temperature data is. The degree of disturbance of the temperature data is smaller as the possibility of abnormality of the temperature data is smaller, and the temperature data should be given a larger weight, and at this time, the local density corresponding to the temperature data is weighted larger. The greater the local reachable density of the temperature data, the denser the distribution of other temperature data around the temperature data, the greater the possibility that the temperature data is data collected under normal conditions, and the greater the weight of the temperature data should be given, and at this time, the greater the weight of the local density corresponding to the temperature data.
And according to the weight of the local density, LOF value of each temperature data is obtained by using an LOF abnormality detection algorithm. In the LOF abnormality detection algorithm, the LOF value is a local outlier factor.
The LOF value of the temperature data is calculated as follows.
In the method, in the process of the invention,Represents the/>Temperature data/>Local outlier factors of (a); /(I)Representation/>Is a local reachable density of (3); Representing a first preset threshold value, and taking an empirical value of 9; /(I) Represents the/>Temperature data/>Included in the neighborhood of (1)A weight of the local density of the individual temperature data; /(I)Represents the/>Temperature data/>Included in the neighborhood of (1)And temperature data.
Thus, LOF values of all temperature data are obtained.
When the LOF value of the temperature data is greater than or equal to the selection threshold value, the temperature data is considered to be abnormal data. Wherein the empirical value of the selection threshold is 1.
When abnormal data are detected, the double-disc motor bearing is considered to have potential problems, and the double-disc motor bearing needs to be overhauled in time. When no abnormal data is detected, the double-disc motor bearing is considered to have no potential problem and is in a normal running state.
Thus, intelligent analysis of the operation data of the double-disc motor bearing is completed.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. The intelligent analysis method for the operation data of the double-disc motor bearing is characterized by comprising the following steps of:
Acquiring temperature data of a motor, preprocessing the temperature data, and acquiring rotating speed data corresponding to the temperature data;
establishing a temperature data set according to the temperature data, acquiring the neighbor radius of the temperature data, determining the neighbor temperature value and the local reachable density of the temperature data, and acquiring the abnormal possibility of the temperature data according to the neighbor radius, the neighbor temperature value and the local reachable density of the temperature data in the temperature data set;
Establishing a temperature-rotating speed plane rectangular coordinate system, acquiring a linear relation corresponding to temperature data and rotating speed data, acquiring a fitting value of each temperature data, establishing a first comparison straight line and vectors corresponding to the corresponding temperature data and rotating speed data, determining a deviation included angle of the corresponding temperature data and rotating speed data, and acquiring the credibility of the temperature data according to the abnormal possibility of the temperature data contained in the adjacent radius of the temperature data, the deviation included angle of the temperature data and the rotating speed data contained in the adjacent radius of the temperature data and the fitting value of the temperature data;
And determining the weight of the local density in the LOF value acquisition process according to the credibility and the abnormal possibility of the temperature data, acquiring the LOF value of each temperature data according to the weight of the local density, and completing intelligent analysis of the running data of the double-disk motor bearing according to the LOF value of the temperature data.
2. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 1, wherein the establishing a temperature data set according to the temperature data comprises the following specific steps: the set of temperature data at all acquisition times is denoted as a temperature dataset.
3. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 1, wherein the method for acquiring the neighbor radius of the temperature data comprises the following specific steps:
And respectively marking each temperature data as temperature data to be analyzed, marking first preset threshold value data closest to the numerical value of the temperature data to be analyzed as neighbor edge data of the temperature data to be analyzed, and marking the absolute value of the maximum value of the difference value between the neighbor edge data of the temperature data to be analyzed and the temperature data to be analyzed as the neighbor radius of the temperature data to be analyzed.
4. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 3, wherein the determining of the neighbor temperature value and the local reachable density of the temperature data comprises the following specific steps:
and recording the average value of all the temperature data contained in the neighbor radius of the temperature data to be analyzed as the neighbor temperature value of the temperature data to be analyzed.
5. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 1, wherein the establishment of the temperature-rotating speed plane rectangular coordinate system comprises the following specific steps: and establishing a temperature-rotating speed plane rectangular coordinate system by taking the temperature data as a vertical axis and the rotating speed data as a horizontal axis.
6. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 1, wherein the linear relation corresponding to the obtained temperature data and the rotation speed data comprises the following specific steps: and performing linear fitting by taking the temperature data as a dependent variable and the rotating speed data as an independent variable to obtain a linear relation corresponding to the temperature data and the rotating speed data.
7. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 1, wherein the method for obtaining the first comparative straight line is as follows: in a rectangular coordinate system of a temperature-rotating speed plane, a straight line corresponding to a linear relation between temperature data and rotating speed data is recorded as a first comparison straight line.
8. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 1, wherein the determining the deviation included angle of the corresponding temperature data and the corresponding rotation speed data comprises the following specific steps:
And recording the included angle between the vector corresponding to the corresponding temperature data and the rotating speed data and the first comparison straight line as the deviated included angle of the group of corresponding temperature data and rotating speed data.
9. The intelligent analysis method for the operation data of the double-disk motor bearing according to claim 1, wherein the determining the weight of the local density in the LOF value obtaining process according to the credibility and the abnormal possibility of the temperature data, and obtaining the LOF value of each temperature data according to the weight of the local density comprises the following specific steps:
In the method, in the process of the invention, Represents the/>Temperature data/>In the LOF value acquisition process, the weight of the local density; /(I)Represents the/>Temperature data/>Is the degree of confidence of (2); /(I)Represents the/>Temperature data/>Is an abnormal possibility of (1); /(I)Representing a first adjustment parameter; /(I)The representation contains the/>Neighborhood number of individual temperature data; /(I)The representation contains the/>First/>, of the temperature dataThe credibility of the temperature data corresponding to each neighborhood; /(I)The representation contains the/>First/>, of the temperature dataLocal reachable density of temperature data corresponding to each neighborhood;
In the method, in the process of the invention, Represents the/>Temperature data/>The LOF value of (2); /(I)Representation/>Is a local reachable density of (3); /(I)Representing a first preset threshold; /(I)Represents the/>Temperature data/>Included in the neighborhood of (1)A weight of the local density of the individual temperature data; represents the/> Temperature data/>Included in the neighborhood of (1)And temperature data.
10. The intelligent analysis method for the operation data of the double-disc motor bearing according to claim 1, wherein the intelligent analysis for the operation data of the double-disc motor bearing is completed according to the LOF value of the temperature data, and the specific method comprises the following steps:
when the LOF value of the temperature data is larger than or equal to the selection threshold value, the temperature data is considered to be abnormal data;
when abnormal data are detected, the double-disc motor bearing is considered to have potential problems, the double-disc motor bearing needs to be overhauled in time, and otherwise, the double-disc motor bearing is considered to operate normally.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410303752.9A CN117909692B (en) | 2024-03-18 | Intelligent analysis method for operation data of double-disc motor bearing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410303752.9A CN117909692B (en) | 2024-03-18 | Intelligent analysis method for operation data of double-disc motor bearing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117909692A true CN117909692A (en) | 2024-04-19 |
| CN117909692B CN117909692B (en) | 2024-05-31 |
Family
ID=
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102706563A (en) * | 2012-06-14 | 2012-10-03 | 哈尔滨工业大学 | Detection method for neighbor abnormities of gas turbine |
| JP2016099863A (en) * | 2014-11-25 | 2016-05-30 | 日本電信電話株式会社 | Abnormal value detection device and operation method thereof |
| CN106053090A (en) * | 2016-08-22 | 2016-10-26 | 中国船舶重工集团公司第七�三研究所 | Neighbor abnormal detection system of gas turbine |
| CN110410282A (en) * | 2019-07-24 | 2019-11-05 | 河北工业大学 | Wind turbines health status on-line monitoring and method for diagnosing faults based on SOM-MQE and SFCM |
| CN113158871A (en) * | 2021-04-15 | 2021-07-23 | 重庆大学 | Wireless signal intensity abnormity detection method based on density core |
| CN113236508A (en) * | 2021-05-31 | 2021-08-10 | 浙江运达风电股份有限公司 | Method for detecting wind speed-power abnormal data of wind generating set |
| WO2022113501A1 (en) * | 2020-11-26 | 2022-06-02 | 三菱重工業株式会社 | Anomaly detection system, anomaly detection method, and program |
| US20220284076A1 (en) * | 2021-03-04 | 2022-09-08 | Korea Advanced Institute Of Science And Technology | Real-time outlier detection method and apparatus in multidimensional data stream |
| CN115238223A (en) * | 2022-09-23 | 2022-10-25 | 江苏泰恩特环境技术有限公司 | Health detection method and system for air-cooled modular machine |
| CN116257402A (en) * | 2022-12-15 | 2023-06-13 | 中国工商银行股份有限公司 | Abnormal data detection method and device, electronic equipment and storage medium |
| US20230191476A1 (en) * | 2021-12-20 | 2023-06-22 | Aktiebolaget Skf | Real-time monitoring method and stability analysis method for continuous casting process |
| CN117272214A (en) * | 2023-11-21 | 2023-12-22 | 罗普特科技集团股份有限公司 | Intelligent management system for safe production |
| CN117436024A (en) * | 2023-12-19 | 2024-01-23 | 湖南翰文云机电设备有限公司 | Fault diagnosis method and system based on drilling machine operation data analysis |
| CN117573477A (en) * | 2023-11-28 | 2024-02-20 | 中国工商银行股份有限公司 | Abnormal data monitoring method, device, equipment, medium and program product |
| US20240085866A1 (en) * | 2022-09-14 | 2024-03-14 | University Of Science And Technology Beijing | Method and device for intelligent control of heating furnace combustion based on a big data cloud platform |
| CN117708748A (en) * | 2024-02-05 | 2024-03-15 | 苏州众志新环冷却设备有限公司 | Operation monitoring system and method for centrifugal fan |
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102706563A (en) * | 2012-06-14 | 2012-10-03 | 哈尔滨工业大学 | Detection method for neighbor abnormities of gas turbine |
| JP2016099863A (en) * | 2014-11-25 | 2016-05-30 | 日本電信電話株式会社 | Abnormal value detection device and operation method thereof |
| CN106053090A (en) * | 2016-08-22 | 2016-10-26 | 中国船舶重工集团公司第七�三研究所 | Neighbor abnormal detection system of gas turbine |
| CN110410282A (en) * | 2019-07-24 | 2019-11-05 | 河北工业大学 | Wind turbines health status on-line monitoring and method for diagnosing faults based on SOM-MQE and SFCM |
| WO2022113501A1 (en) * | 2020-11-26 | 2022-06-02 | 三菱重工業株式会社 | Anomaly detection system, anomaly detection method, and program |
| US20220284076A1 (en) * | 2021-03-04 | 2022-09-08 | Korea Advanced Institute Of Science And Technology | Real-time outlier detection method and apparatus in multidimensional data stream |
| CN113158871A (en) * | 2021-04-15 | 2021-07-23 | 重庆大学 | Wireless signal intensity abnormity detection method based on density core |
| CN113236508A (en) * | 2021-05-31 | 2021-08-10 | 浙江运达风电股份有限公司 | Method for detecting wind speed-power abnormal data of wind generating set |
| US20230191476A1 (en) * | 2021-12-20 | 2023-06-22 | Aktiebolaget Skf | Real-time monitoring method and stability analysis method for continuous casting process |
| US20240085866A1 (en) * | 2022-09-14 | 2024-03-14 | University Of Science And Technology Beijing | Method and device for intelligent control of heating furnace combustion based on a big data cloud platform |
| CN115238223A (en) * | 2022-09-23 | 2022-10-25 | 江苏泰恩特环境技术有限公司 | Health detection method and system for air-cooled modular machine |
| CN116257402A (en) * | 2022-12-15 | 2023-06-13 | 中国工商银行股份有限公司 | Abnormal data detection method and device, electronic equipment and storage medium |
| CN117272214A (en) * | 2023-11-21 | 2023-12-22 | 罗普特科技集团股份有限公司 | Intelligent management system for safe production |
| CN117573477A (en) * | 2023-11-28 | 2024-02-20 | 中国工商银行股份有限公司 | Abnormal data monitoring method, device, equipment, medium and program product |
| CN117436024A (en) * | 2023-12-19 | 2024-01-23 | 湖南翰文云机电设备有限公司 | Fault diagnosis method and system based on drilling machine operation data analysis |
| CN117708748A (en) * | 2024-02-05 | 2024-03-15 | 苏州众志新环冷却设备有限公司 | Operation monitoring system and method for centrifugal fan |
Non-Patent Citations (4)
| Title |
|---|
| JIABIN LIU，ETC: "Self-Supervised Random Forest on Transformed Distribution for Anomaly Detection", IEEE, 23 January 2024 (2024-01-23) * |
| 周鹏;程艳云;: "一种改进的LOF异常点检测算法", 计算机技术与发展, no. 12, 10 December 2017 (2017-12-10) * |
| 邹云峰;张昕;宋世渊;倪巍伟;: "基于局部密度的快速离群点检测算法", 计算机应用, no. 10, 10 October 2017 (2017-10-10) * |
| 马斌彬: "基于性能退化分析的风电机组关键设计参数识别", 中国优秀硕士论文电子期刊网, 15 September 2021 (2021-09-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353482A (en) | LSTM-based fatigue factor recessive anomaly detection and fault diagnosis method | |
| JP6887361B2 (en) | Monitoring target selection device, monitoring target selection method, and program | |
| CN111178327A (en) | Deep learning-based bearing state identification method and system | |
| CN116304848B (en) | Rolling bearing fault diagnosis system and method | |
| CN113340598B (en) | Rolling bearing intelligent fault diagnosis method based on regularized sparse model | |
| US11555757B2 (en) | Monitoring device, monitoring method, method of creating shaft vibration determination model, and program | |
| CN117454283A (en) | State evaluation method for wind turbine generator operation detection data | |
| CN117629615B (en) | Shield tunneling machine fault detection method, equipment and medium | |
| CN117436712B (en) | Real-time monitoring method and system for operation risk of construction hanging basket | |
| JP6783163B2 (en) | Wing vibration monitoring device and wing vibration monitoring method | |
| CN117909692B (en) | Intelligent analysis method for operation data of double-disc motor bearing | |
| Atta et al. | Detection and diagnosis of bearing faults under fixed and time-varying speed conditions using persistence spectrum and multi-scale structural similarity index | |
| CN114088389A (en) | Data processing method and related device for gearbox | |
| CN117909692A (en) | Intelligent analysis method for operation data of double-disc motor bearing | |
| CN111797545A (en) | Wind turbine generator yaw reduction coefficient calculation method based on measured data | |
| Ma et al. | Research on condition monitoring of bearing health using vibration data | |
| Lv et al. | General log-linear weibull model combining vibration and temperature characteristics for remaining useful life prediction of rolling element bearings | |
| EP3819608A1 (en) | Detecting rotor anomalies by determining vibration trends during transient speed operation | |
| KR20230100999A (en) | Method for determining dominant frequency of rotating machinery facilities, and computing system performing the same | |
| CN115356109B (en) | Rolling bearing fault identification method and system | |
| CN117591970B (en) | Fault early warning method and system for full-automatic assembly equipment | |
| CN116244622A (en) | K-nearest neighbor centrifugal pump fault diagnosis method based on improved Relieff weighting | |
| CN116256589B (en) | Intelligent diagnosis method and device for electric pump well, storage medium and server | |
| Liu et al. | Condition Monitoring Method of Mechanical Equipment Based on Data Boundary Morphology | |
| KR102231315B1 (en) | Method for diagnosis bearing fault based on band optimization and apparatus for performing the same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |