CN117349711B - Electronic tag data processing method and system for railway locomotive parts - Google Patents
Electronic tag data processing method and system for railway locomotive parts Download PDFInfo
- Publication number
- CN117349711B CN117349711B CN202311640888.0A CN202311640888A CN117349711B CN 117349711 B CN117349711 B CN 117349711B CN 202311640888 A CN202311640888 A CN 202311640888A CN 117349711 B CN117349711 B CN 117349711B
- Authority
- CN
- China
- Prior art keywords
- data
- electronic tag
- monitoring index
- value
- parts
- 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.)
- Active
Links
- 230000003137 locomotive effect Effects 0.000 title claims abstract description 45
- 238000003672 processing method Methods 0.000 title claims description 7
- 230000002159 abnormal effect Effects 0.000 claims abstract description 67
- 238000000034 method Methods 0.000 claims abstract description 54
- 230000005856 abnormality Effects 0.000 claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000012544 monitoring process Methods 0.000 claims description 192
- 238000007689 inspection Methods 0.000 claims description 141
- 230000008569 process Effects 0.000 claims description 13
- 238000001514 detection method Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 description 9
- 230000032683 aging Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 239000006096 absorbing agent Substances 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24143—Distances to neighbourhood prototypes, e.g. restricted Coulomb energy networks [RCEN]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a method and a system for processing electronic tag data of railway locomotive parts, comprising the following steps: acquiring electronic tag data of a locomotive part, acquiring a state characteristic value of any part according to the electronic tag data, acquiring state characteristic value data of any part according to the electronic tag data and the state characteristic value of any part, acquiring the abnormality degree of any one of the electronic tag data according to the state characteristic value data of any part, acquiring initial abnormality data according to the abnormality degree, and acquiring abnormal parts in the locomotive part according to the initial abnormality data. According to the method and the device, the optimal K value is obtained according to the obtained data, so that abnormal data of the data can be detected more accurately, abnormal parts are obtained, the abnormal parts comprise possibly abnormal parts, and the prediction of the abnormality of the parts is realized.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a system for processing electronic tag data of railway locomotive parts.
Background
The locomotive state in railway transportation directly affects the running efficiency and the safety of a railway system, so that effective abnormal data detection on locomotive parts is very critical, in the existing algorithm, a plurality of methods for identifying abnormal data are provided, wherein the LOF algorithm can accurately identify the abnormal data, and the basic idea of the LOF algorithm is to detect the abnormality by comparing the local density difference of an object and the neighbors of the object. For each data point, the LOF algorithm calculates the density of other points in its neighborhood and then compares this density to the density of its neighbors. If the density of a data point is much lower than the density of its neighbors, then that point is considered likely to be an outlier. However, when the algorithm acquires abnormal data points, the selection of the number K of neighbors is very critical, and the K value can be difficult to select due to the numerical variation of various monitoring indexes of the parts during the multi-time inspection of the electronic tag data of the parts of the railway locomotive, and the optimal values of different data sets can be different.
Disclosure of Invention
In order to solve the problems, the invention provides a method and a system for processing electronic tag data of railway locomotive parts.
The invention relates to a method and a system for processing electronic tag data of railway locomotive parts, which adopts the following technical scheme:
the invention provides a method for processing electronic tag data of railway locomotive parts, which comprises the following steps:
acquiring electronic tag data of railway locomotive parts, wherein the electronic tag data comprises a plurality of inspection data, and each inspection data comprises numerical values of various monitoring indexes of the plurality of parts;
obtaining a state characteristic value of any part according to the numerical value of different monitoring indexes of any part in the electronic tag data;
any one monitoring index of any one part in the inspection data corresponds to one data point, wherein the abscissa of the data point is a state characteristic value of the part, and the ordinate is a numerical value of any one monitoring index of the part;
for all parts of the same part in any one-time inspection data, taking all data points of all parts under any one monitoring index as state characteristic value data of any one part in any one monitoring index in any one-time inspection data, wherein the part is a plurality of parts with the same name;
obtaining the abnormal degree of any one monitoring index of any one part in any one time of inspection data according to the monitoring index of the data point in the state characteristic value data of any one monitoring index of any one part in any one time of inspection data and the monitoring index of the corresponding part of the electronic tag data, and obtaining all initial abnormal data in the electronic tag data according to the abnormal degree of any one monitoring index of any one part in any one time of inspection data in the electronic tag data;
obtaining the K value of each monitoring index of each part in each inspection data in the electronic tag data according to all initial abnormal data in the electronic tag data, wherein the K value is the K value of an LOF algorithm, and abnormal parts in locomotive parts are obtained according to the K value of each monitoring index of each part in each inspection data in the electronic tag data.
Further, the method for obtaining the state characteristic value of any part according to the numerical value of different monitoring indexes of any part in the electronic tag data comprises the following specific steps:
and obtaining the state characteristic value of any part according to the ratio of the numerical value of different monitoring indexes of any part in the electronic tag data to the maximum value of the monitoring indexes.
Further, the method obtains the state characteristic value of any part according to the ratio of the numerical value of different monitoring indexes to the maximum value of the monitoring index of any part in the electronic tag data, and comprises the following specific steps:
in the method, in the process of the invention,the ith inspection data representing the electronic tag data>The parts are at the (th)>Numerical value of seed monitoring index,/->Representing the +.>The parts are at the (th)>Maximum value of species monitoring index,/->Indicating the total number of inspection data in the electronic tag data, +.>Indicating the total number of categories of monitoring indicators of the component, < ->Indicate->Status characteristic values of the individual components.
Further, the specific method for obtaining the abnormality degree is as follows:
and obtaining the abnormality degree of any one monitoring index of any one part in the electronic tag data according to the average value difference between the monitoring index of the data point in the state characteristic value data of any one monitoring index of any one part in the inspection data and the monitoring index of all the data points and the monitoring index of the corresponding part of the electronic tag data.
Further, according to the average difference between the monitoring index of the data point in the state characteristic value data of any one of the monitoring indexes of any one of the parts in the inspection data at any one time and the monitoring index of all the data points and the monitoring index of the corresponding part of the electronic tag data, the abnormal degree of any one of the monitoring indexes of any one of the parts in the inspection data at any one time in the electronic tag data is obtained, which comprises the following specific steps:
in the method, in the process of the invention,indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Anomaly coefficient of data point, +.>Indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>The amplitude of the data point, wherein the amplitude represents the value of the monitoring index corresponding to the data point, +.>Indicating +.f in the ith inspection data>The average value of the amplitude values of all data points of the j-th monitoring index state characteristic value data of the component>Indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Slope of data point, +.>Taking an absolute value;
representing a linear normalization function, ++>The ith inspection data representing the electronic tag data>The parts are at the (th)>The value of the species monitoring index, wherein->The parts are->The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Parts corresponding to data points, < >>Indicating the +.f in all the inspection data>No. H of the parts>Minimum value of species monitoring index, +.>Indicating the +.f in all the inspection data>All of the parts->Variance of the values of the species monitoring index, +.>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The degree of abnormality of the index is monitored.
Further, the method obtains all initial abnormal data in the electronic tag data according to the abnormal degree of any monitoring index of any part in the inspection data at any time in the electronic tag data, and comprises the following specific steps:
indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The degree of numerical abnormality of the species monitoring index, when +.>When the electronic tag is used, the (th) in the ith inspection data in the electronic tag data is added>No. H of the parts>The numerical value of the seed monitoring index is used as initial abnormal data, and all initial abnormal data in the electronic tag data are acquired>Is a preset first value.
Further, the method for obtaining the K value of each monitoring index of each part in each inspection data in the electronic tag data according to all initial abnormal data in the electronic tag data comprises the following specific steps:
and obtaining the K value of the numerical value of each monitoring index of each part in each inspection data in the electronic tag data according to the distance difference between the numerical value of any monitoring index of any part in the inspection data at any time in the electronic tag data and the latest initial abnormal data.
Further, the method for obtaining the K value of the value of each monitoring index of each component in each inspection data in the electronic tag data according to the distance difference between the value of any monitoring index of any component in the inspection data at any time in the electronic tag data and the latest initial abnormal data comprises the following specific steps:
the method comprises the steps of arranging the numerical values of all monitoring indexes of all parts in all inspection data in electronic tag data according to an acquisition time sequence to obtain a numerical sequence of the monitoring indexes, marking the numerical value of any one monitoring index of any part in any one inspection data in the electronic tag data as TS1 in the numerical sequence, obtaining the numerical value of initial abnormal data nearest to TS1 in the numerical sequence as TS2, and recording the numerical value of any one monitoring index of any part in any one inspection data in the electronic tag dataAs the distance difference between the numerical value of any monitoring index of any part in any one of the inspection data in the electronic tag data and the latest initial abnormal data;
in the method, in the process of the invention,indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The difference of the distance between the value of the seed monitoring index and the last initial abnormal data,/for the seed monitoring index>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The variance of the values of all monitoring indicators between the values of the monitoring indicators and the last initial anomaly data,/>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>Numerical abnormality degree of species monitoring index, +.>As an exponential function based on natural constants, < +.>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>K value of the numerical value of the seed monitoring index;
and acquiring the K value of the numerical value of each monitoring index of each part in each inspection data in the electronic tag data.
Further, the method comprises the following specific steps of:
and carrying out anomaly detection on the numerical value of each monitoring index of each part in each inspection data in the electronic tag data by using an LOF algorithm according to the K value to obtain all anomaly data in the electronic tag data, and marking the parts corresponding to all anomaly data to obtain the anomaly parts in locomotive parts.
The invention also provides a railway locomotive part electronic tag data processing system which comprises a memory and a processor, wherein the processor executes a computer program stored in the memory so as to realize the steps of the method.
The technical scheme of the invention has the beneficial effects that: when the electronic tag data of the railway locomotive parts are processed, the states of the parts are required to be evaluated according to the inspection data of each time, so that the electronic tag data are subjected to anomaly detection through the LOF algorithm, and the K value is related to the accuracy of data anomaly detection.
When the K value is self-adaptive according to the change of the data, the state characteristic value of any part is obtained according to the electronic tag data, the state characteristic value can reflect the state information of the part in the process of multi-time inspection, the state characteristic value data of any part is obtained according to the electronic tag data and the state characteristic value of any part, the state characteristic value data analyzes the state information of the same part, the abnormality degree of any part in the electronic tag data is obtained according to the state characteristic value data of any part and the electronic tag data, all the abnormality data in the electronic tag data are obtained according to the abnormality degree, and then the marked part possibly has no substantial quality problem by marking the part corresponding to the abnormality data, but the marked part possibly has the quality problem along with the increase of the use time, and the prediction of the abnormality of the locomotive part is achieved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of a method for processing electronic tag data of a component of a railroad locomotive according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of the specific implementation, structure, characteristics and effects of the electronic tag data processing method and system for railway locomotive components according to the invention, which are provided by the invention, with reference to the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a concrete scheme of a railway locomotive part electronic tag data processing method, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for processing electronic tag data of a railway locomotive component according to an embodiment of the present invention is shown, the method includes the following steps:
and S001, acquiring electronic tag data of railway locomotive parts.
It should be noted that, the main purpose of this embodiment is to detect abnormal data in electronic tag data of a railway locomotive component, so that it is first necessary to acquire required electronic tag data before the detection is performed.
Specifically, electronic tag data of the railway locomotive component is obtained through a background management system for storing information of the railway locomotive component, wherein the electronic tag data is one-dimensional data and comprises multiple inspection data, each inspection data comprises numerical values of various monitoring indexes of the multiple components, and the multiple components include but are not limited to: engines, drive shafts, gearboxes, suspension springs, shock absorbers, air brake systems, electric brake systems, and the like, with various monitoring indicators including, but not limited to: the monitoring indexes are text information recorded by detection personnel, such as the running condition of each part, whether cracks, abrasion, abnormal sounds exist or not, and the like.
It should be noted that, because the electronic tag data includes multiple inspection data, and text information exists in the monitoring index information of each part in each inspection data, the subsequent analysis in this embodiment is mainly numerical information, so the text information in the electronic tag data needs to be converted.
Further, the TF-IDF algorithm is utilized to convert the Chinese information in the inspection data of the electronic tag data into numerical information.
It should be noted that, in the present method of converting the Chinese information in the inspection data of the electronic tag data into the numerical information as the TF-IDF algorithm, the embodiment will not be described again, and the processed electronic tag data is still recorded as the electronic tag data of the locomotive component.
Thus, the electronic tag data of the railway locomotive parts are obtained.
Step S002, obtaining the state characteristic value of any part according to the numerical value of different monitoring indexes of any part in the electronic tag data.
It should be noted that, since there are many identical parts in the locomotive parts, the aging of the parts is a slow changing process, if a substantial quality problem occurs in a certain part, it can be clearly observed, and in each inspection process, all the main parts need to be inspected, so the state information of all the main parts is updated each time, and the indexes such as the aging degree and the like of the parts in the same batch should be similar, so when the state information of the parts is evaluated, it is necessary to analyze not only the change of the data of each part, but also the change of the data of the parts of the same kind, so as to obtain the state information which may have abnormality, so the state detection of the parts is more accurate, even if the substantial quality problem does not occur in the parts, but as the use time increases, the state information may occur, so when the state information is detected, it should be regarded as the abnormal data.
It should be further noted that, in this embodiment, the LOF algorithm is selected to detect abnormal data that may exist in the monitoring data of the component, and when the LOF algorithm is used to detect abnormal data that may exist in the monitoring data of the component, the algorithm detects the abnormality by comparing the local density difference between an object and its neighboring. For each data point, the LOF algorithm calculates the density of other points in its neighborhood and then compares this density to the density of its neighbors. If the density of a data point is much lower than the density of its neighbors, then that point is considered likely to be an outlier. However, when the algorithm acquires abnormal data points, the selection of the number K of neighbors is very critical, so the embodiment adaptively acquires the K value according to the data change.
Specifically, the state characteristic value of any part is obtained according to the numerical value of different monitoring indexes of any part in the electronic tag data, and the state characteristic value is specifically as follows:
in the method, in the process of the invention,the ith inspection data representing the electronic tag data>The parts are at the (th)>Numerical value of seed monitoring index,/->Representing the +.>The parts are at the (th)>Maximum value of species monitoring index,/->Indicating the total number of inspection data in the electronic tag data, +.>Indicating the total number of categories of monitoring indicators of the component, < ->Indicate->Status characteristic values of the individual components.
It should be noted that the number of the substrates,the value of each monitoring data is normalized, and the state characteristic value represents the change of the whole state data of the part in the process of the previous inspection.
Thus, the state characteristic value of any part is obtained.
Step S003, according to the state characteristic value of any part in any one inspection data and the state characteristic value data of any part in any one inspection data, according to the monitoring index of the data point of any part in any one inspection data and the monitoring index of the part corresponding to the electronic tag data, obtaining the abnormality degree of any one inspection index of any part in any one inspection data in the electronic tag data, and according to the abnormality degree of any one monitoring index of any part in any one inspection data in the electronic tag data, obtaining all initial abnormality data in the electronic tag data.
It should be noted that, since the data changes of the components of the same kind need to be analyzed to obtain the state information that may have an abnormality, so that the state detection of the components is more accurate, it is necessary to classify all the components in the electronic tag data and analyze each kind.
Specifically, according to the electronic tag data and the state characteristic value data of any one of the parts in any one monitoring index in any one of the inspection data, the method specifically comprises the following steps:
any one monitoring index of any one part in the inspection data corresponds to a data point, wherein the abscissa of the data point is a state characteristic value of the part, and the ordinate is a numerical value of any one monitoring index of the part.
And regarding all the parts of the same part in the random inspection data, taking all data points of all the parts under any monitoring index as state characteristic value data of any part in any monitoring index in the random inspection data, wherein any part is a plurality of parts with the same name.
The state characteristic value data of any one of the components is obtained, and the abnormality degree of the numerical value of the monitoring index in the electronic tag data is analyzed through the amplitude change of different data points in the state characteristic value data.
Specifically, according to the monitoring index of the data point of any part in the state characteristic value data of any monitoring index in any one time of inspection data and the monitoring index of the part corresponding to the electronic tag data, the abnormal degree of any monitoring index of any part in any one time of inspection data in the electronic tag data is obtained, specifically as follows:
in the method, in the process of the invention,indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Anomaly coefficient of data point, +.>Indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>The amplitude of the data point, wherein the amplitude represents the value of the monitoring index corresponding to the data point, +.>Indicating +.f in the ith inspection data>The average value of the amplitude values of all data points of the j-th monitoring index state characteristic value data of the component>Indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Slope of data point, +.>To take absolute value.
Representing a linear normalization function, the normalized object being +.>，The ith inspection data representing the electronic tag data>The parts are at the (th)>The value of the species monitoring index, wherein->The parts are->The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Parts corresponding to data points, < >>Indicating the +.f in all the inspection data>No. H of the parts>Minimum value of species monitoring index, +.>Indicating the +.f in all the inspection data>All of the parts->Variance of the values of the species monitoring index, +.>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The degree of abnormality of the index is monitored.
It should be noted that, the slope of the data point in the state characteristic value data may be obtained by fitting the state characteristic value data to a polynomial curve of five times by using a least square method, which is not described in detail in this embodiment.
It should be noted that the number of the substrates,indicate->Status feature value data of the seed component +.>The greater the difference between the amplitude and the mean of the data points, the more pronounced +.>The more likely an anomaly is present in a data point, multiplied by the slope +.>Indicate->The larger the slope of the change in data point, the greater the data difference between the data and its neighborhood data point, and therefore the more likely it is that an anomaly is present. />The ith inspection data representing the electronic tag data>No. H of the parts>The ratio of the value of the monitoring index to the minimum value, where the minimum value is taken as a reference data point, the greater the ratio, the more ∈>The greater the degree of numerical variation of a data point, and therefore the more likely it is that an anomaly is present, the greater the variance is indicative of the overall degree of variation of the data, the more likely it is that an anomaly is present for such data, the greater is the variance is, and the multiplication is +.>The weight coefficient is represented because the value represents the degree of abnormality of each component in the same type of component.
It should be noted that, the value abnormality degree of any one monitoring index of any one component in any one inspection data in the electronic tag data is obtained, and all initial abnormality data in the electronic tag data is obtained through threshold value screening, so that the K value of the value of each monitoring index in the electronic tag data is conveniently and subsequently determined, and further accurate abnormality data in the electronic tag data is determined.
Specifically, all initial abnormal data in the electronic tag data are obtained according to the abnormal degree of any monitoring index of any part in any one time of inspection data in the electronic tag data, and the method specifically comprises the following steps:
when (when)When the electronic tag is used, the (th) in the ith inspection data in the electronic tag data is added>No. H of the parts>The numerical value of the seed monitoring index is used as initial abnormal data, and all initial abnormal data in the electronic tag data are acquired>To preset the first value, the present embodiment uses +.>Examples are described.
So far, all initial abnormal data in the electronic tag data are obtained.
Step S004, obtaining the K value of each monitoring index of each part in each inspection data in the electronic tag data according to all initial abnormal data in the electronic tag data, and obtaining abnormal parts in locomotive parts according to the K value of each monitoring index of each part in each inspection data in the electronic tag data.
It should be noted that, the initial anomaly data obtained above is then used to determine the K value according to the change of the neighborhood data of the data, where in the LOF algorithm, the K value represents the size of the local neighborhood, which may also be referred to as the number of K neighbors, and for each data, the LOF algorithm calculates the density ratio of the data to its K nearest neighbors to determine whether it is anomaly. When the K value is smaller, the algorithm is more sensitive, points with smaller local density differences are easily marked as abnormal values, and excessive data can be wrongly marked as the abnormal values to generate false positives; when the K value is large, the algorithm will consider a broader neighborhood, marking the points as normal more easily, even though they have some density differences in the local area, which may lead to false negatives, i.e. true outliers are not detected, thus determining the K value for each data from the change in data.
Specifically, the K value of each monitoring index of each part in each inspection data in the electronic tag data is obtained according to all initial abnormal data in the electronic tag data, and the K value is specifically as follows:
the method comprises the steps of arranging the numerical values of all monitoring indexes of all parts in all inspection data in electronic tag data according to an acquisition time sequence to obtain a numerical sequence of the monitoring indexes, marking the numerical value of any one monitoring index of any part in any one inspection data in the electronic tag data as TS1 in the numerical sequence, obtaining the numerical value of initial abnormal data nearest to TS1 in the numerical sequence as TS2, and recording the numerical value of any one monitoring index of any part in any one inspection data in the electronic tag dataAnd the difference of the distance between the numerical value of any monitoring index of any part in any one inspection data in the electronic tag data and the latest initial abnormal data is used as the distance difference between any one monitoring index of any part in any one inspection data in the electronic tag data and the latest initial abnormal data.
In the method, in the process of the invention,indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The difference of the distance between the value of the seed monitoring index and the last initial abnormal data,/for the seed monitoring index>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The variance of the values of all monitoring indicators between the values of the monitoring indicators and the last initial anomaly data,/>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>Numerical abnormality degree of species monitoring index, +.>As an exponential function based on natural constants, < +.>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>K value of the numerical value of the monitoring index.
And acquiring the K value of the numerical value of each monitoring index of each part in each inspection data in the electronic tag data.
The variance indicates the fluctuation of the neighborhood data, and the larger the fluctuation degree is, the larger the variation degree of the specification data is, and when abnormality of the data is detected, the smaller the neighborhood data is required to describe the abnormality thereof, so the larger the variance is, the smaller the K value required for the specification of the data is. Multiplying by the degree of anomaly of the data, the greater the degree of anomaly, the greater the degree of variation of the data points, and therefore, in order to be able to better detect it, a smaller neighborhood radius is required.
Further, according to the K value of each monitoring index of each part in each inspection data in the electronic tag data, the abnormal part in the locomotive part is specifically as follows:
and carrying out anomaly detection on the numerical value of each monitoring index of each part in each inspection data in the electronic tag data by using an LOF algorithm according to the K value to obtain all anomaly data in the electronic tag data, marking the parts corresponding to all anomaly data to obtain the anomaly parts in locomotive parts, and subsequently replacing or overhauling the anomaly parts. It should be noted that, the LOF algorithm is a prior art, and will not be described in detail herein, and even if no substantial quality problem occurs in the corresponding parts in all the obtained abnormal data, the quality problem may occur with the increase of the usage time, that is, the prediction of the abnormality of the locomotive parts is achieved.
Through the steps, the electronic tag data processing method for the railway locomotive parts is completed.
Another embodiment of the present invention provides a railroad locomotive component electronic label data processing system comprising a memory and a processor, the processor executing a computer program stored in the memory, performing the following operations:
acquiring electronic tag data of railway locomotive parts, wherein the electronic tag data comprises a plurality of inspection data, and each inspection data comprises numerical values of various monitoring indexes of the plurality of parts; obtaining a state characteristic value of any part according to the numerical value of different monitoring indexes of any part in the electronic tag data; any one monitoring index of any one part in the inspection data corresponds to one data point, wherein the abscissa of the data point is a state characteristic value of the part, and the ordinate is a numerical value of any one monitoring index of the part; for all parts of the same part in any one-time inspection data, taking all data points of all parts under any one monitoring index as state characteristic value data of any one part in any one monitoring index in any one-time inspection data, wherein the part is a plurality of parts with the same name; obtaining the abnormal degree of any one monitoring index of any one part in any one time of inspection data according to the monitoring index of the data point in the state characteristic value data of any one monitoring index of any one part in any one time of inspection data and the monitoring index of the corresponding part of the electronic tag data, and obtaining all initial abnormal data in the electronic tag data according to the abnormal degree of any one monitoring index of any one part in any one time of inspection data in the electronic tag data; obtaining the K value of each monitoring index of each part in each inspection data in the electronic tag data according to all initial abnormal data in the electronic tag data, wherein the K value is the K value of an LOF algorithm, and abnormal parts in locomotive parts are obtained according to the K value of each monitoring index of each part in each inspection data in the electronic tag data.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (3)
1. The electronic tag data processing method for the railway locomotive parts is characterized by comprising the following steps of:
acquiring electronic tag data of railway locomotive parts, wherein the electronic tag data comprises a plurality of inspection data, and each inspection data comprises numerical values of various monitoring indexes of the plurality of parts;
obtaining a state characteristic value of any part according to the numerical value of different monitoring indexes of any part in the electronic tag data;
any one monitoring index of any one part in the inspection data corresponds to one data point, wherein the abscissa of the data point is a state characteristic value of the part, and the ordinate is a numerical value of any one monitoring index of the part;
for all parts of the same part in any one-time inspection data, taking all data points of all parts under any one monitoring index as state characteristic value data of any one part in any one monitoring index in any one-time inspection data, wherein the part is a plurality of parts with the same name;
obtaining the abnormal degree of any one monitoring index of any one part in any one time of inspection data according to the monitoring index of the data point in the state characteristic value data of any one monitoring index of any one part in any one time of inspection data and the monitoring index of the corresponding part of the electronic tag data, and obtaining all initial abnormal data in the electronic tag data according to the abnormal degree of any one monitoring index of any one part in any one time of inspection data in the electronic tag data;
obtaining a K value of each monitoring index of each part in each inspection data in the electronic tag data according to all initial abnormal data in the electronic tag data, wherein the K value is a K value of an LOF algorithm, and abnormal parts in locomotive parts are obtained according to the K value of each monitoring index of each part in each inspection data in the electronic tag data;
the method for obtaining the state characteristic value of any part according to the numerical value of different monitoring indexes of any part in the electronic tag data comprises the following specific steps:
obtaining a state characteristic value of any part according to the ratio of the numerical value of different monitoring indexes of any part in the electronic tag data to the maximum value of the monitoring indexes;
the state characteristic value of any part is obtained according to the ratio of the numerical value of different monitoring indexes and the maximum value of the monitoring indexes of any part in the electronic tag data, and the method comprises the following specific steps:
in the method, in the process of the invention,the ith inspection data representing the electronic tag data>The parts are at the (th)>Numerical value of seed monitoring index,/->Representing the +.>The parts are at the (th)>Maximum value of species monitoring index,/->Indicating the total number of inspection data in the electronic tag data, +.>Indicating the total number of categories of monitoring indicators of the component, < ->Indicate->Status feature values of the individual components;
the specific acquisition method of the abnormality degree is as follows:
obtaining the abnormality degree of any one monitoring index of any one part in any one time of inspection data in the electronic tag data according to the average value difference between the monitoring index of the data point in the state characteristic value data of any one monitoring index of any one part in the inspection data and the monitoring index of all data points and the monitoring index of the corresponding part of the electronic tag data;
the method comprises the specific steps of obtaining the abnormality degree of any one monitoring index of any one part in any one time of inspection data in electronic tag data according to the average value difference between the monitoring index of any one part in any one time of inspection data in the state characteristic value data of any one monitoring index and the monitoring index of all data points and the monitoring index of the corresponding part of the electronic tag data, wherein the specific steps are as follows:
in the method, in the process of the invention,indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Anomaly coefficient of data point, +.>Indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>The amplitude of the data point, wherein the amplitude represents the value of the monitoring index corresponding to the data point, +.>Indicating +.f in the ith inspection data>The average value of the amplitude values of all data points of the j-th monitoring index state characteristic value data of the component>Indicating +.f in the ith inspection data>The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Slope of data point, +.>Taking an absolute value;
representing a linear normalization function, ++>The ith inspection data representing the electronic tag data>The parts are at the (th)>The value of the species monitoring index, wherein->The parts are->The component is the (j) th component in the state characteristic value data of the j-th monitoring index>Parts corresponding to data points, < >>Indicating the +.f in all the inspection data>No. H of the parts>Minimum value of species monitoring index, +.>Indicating the +.f in all the inspection data>All of the parts->Variance of the values of the species monitoring index, +.>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>Monitoring the abnormality degree of the index;
the method comprises the following specific steps of:
indicating the number of electronic labelsAccording to the ith inspection data +.>No. H of the parts>The degree of numerical abnormality of the species monitoring index, when +.>When the electronic tag is used, the (th) in the ith inspection data in the electronic tag data is added>No. H of the parts>The numerical value of the seed monitoring index is used as initial abnormal data, and all initial abnormal data in the electronic tag data are acquired>Is a preset first numerical value;
the K value of each monitoring index of each part in each inspection data in the electronic tag data is obtained according to all initial abnormal data in the electronic tag data, and the method comprises the following specific steps:
obtaining the K value of the numerical value of each monitoring index of each part in each inspection data in the electronic tag data according to the distance difference between the numerical value of any monitoring index of any part in the inspection data of any time in the electronic tag data and the nearest initial abnormal data;
the method comprises the following specific steps of:
for all patrol data in the electronic tag dataThe values of all monitoring indexes of all parts are arranged according to the acquisition time sequence to obtain a value sequence of the monitoring indexes, for the value of any one monitoring index of any part in any one time of inspection data in electronic tag data, the number of the value of the monitoring index in the value sequence is marked as TS1, the number of initial abnormal data closest to TS1 is obtained in the value sequence and marked as TS2, and the number of the initial abnormal data closest to TS1 is marked as TS2As the distance difference between the numerical value of any monitoring index of any part in any one of the inspection data in the electronic tag data and the latest initial abnormal data;
in the method, in the process of the invention,indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The difference of the distance between the value of the seed monitoring index and the last initial abnormal data,/for the seed monitoring index>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>The variance of the values of all monitoring indicators between the values of the monitoring indicators and the last initial anomaly data,/>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>Numerical abnormality degree of species monitoring index, +.>As an exponential function based on natural constants, < +.>Indicating the (th) in the ith inspection data in the electronic tag data>No. H of the parts>K value of the numerical value of the seed monitoring index;
and acquiring the K value of the numerical value of each monitoring index of each part in each inspection data in the electronic tag data.
2. The method for processing electronic tag data of railway locomotive parts according to claim 1, wherein the abnormal parts in the locomotive parts are obtained according to the K value of each monitoring index of each part in each inspection data in the electronic tag data, comprising the following specific steps:
and carrying out anomaly detection on the numerical value of each monitoring index of each part in each inspection data in the electronic tag data by using an LOF algorithm according to the K value to obtain all anomaly data in the electronic tag data, and marking the parts corresponding to all anomaly data to obtain the anomaly parts in locomotive parts.
3. A railway locomotive component electronic tag data processing system comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the steps of a railway locomotive component electronic tag data processing method as claimed in any one of claims 1-2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311640888.0A CN117349711B (en) | 2023-12-04 | 2023-12-04 | Electronic tag data processing method and system for railway locomotive parts |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311640888.0A CN117349711B (en) | 2023-12-04 | 2023-12-04 | Electronic tag data processing method and system for railway locomotive parts |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117349711A CN117349711A (en) | 2024-01-05 |
| CN117349711B true CN117349711B (en) | 2024-02-13 |
Family
ID=89356032
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311640888.0A Active CN117349711B (en) | 2023-12-04 | 2023-12-04 | Electronic tag data processing method and system for railway locomotive parts |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117349711B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117668684B (en) * | 2024-01-31 | 2024-04-16 | 新风光电子科技股份有限公司 | Power grid electric energy data anomaly detection method based on big data analysis |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111578116A (en) * | 2020-04-26 | 2020-08-25 | 湖南京辙科技有限公司 | Train wheel oil filling hole screw plug and using method thereof |
| CN114492826A (en) * | 2021-11-22 | 2022-05-13 | 杭州电子科技大学 | Unsupervised anomaly detection analysis solution method based on multivariate time sequence flow data |
| CN114648245A (en) * | 2022-04-07 | 2022-06-21 | 南通俊朗智能科技有限公司 | Equipment inspection priority evaluation method and system based on artificial intelligence |
| CN116012579A (en) * | 2023-02-17 | 2023-04-25 | 厦门破云溯星科技有限公司 | Method for detecting abnormal states of parts based on photographed images of intelligent inspection robot of train |
| CN116630327A (en) * | 2023-07-25 | 2023-08-22 | 江苏太湖锅炉股份有限公司 | Boiler state abnormity monitoring system based on thermodynamic diagram |
| CN116665126A (en) * | 2023-05-29 | 2023-08-29 | 华中科技大学 | Robot inspection part defect detection method and application thereof |
| CN116702081A (en) * | 2023-08-07 | 2023-09-05 | 西安格蒂电力有限公司 | Intelligent inspection method for power distribution equipment based on artificial intelligence |
| CN116879662A (en) * | 2023-09-06 | 2023-10-13 | 山东华尚电气有限公司 | Transformer fault detection method based on data analysis |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230049418A1 (en) * | 2021-08-10 | 2023-02-16 | Capital One Services, Llc | Information quality of machine learning model outputs |
-
2023
- 2023-12-04 CN CN202311640888.0A patent/CN117349711B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111578116A (en) * | 2020-04-26 | 2020-08-25 | 湖南京辙科技有限公司 | Train wheel oil filling hole screw plug and using method thereof |
| CN114492826A (en) * | 2021-11-22 | 2022-05-13 | 杭州电子科技大学 | Unsupervised anomaly detection analysis solution method based on multivariate time sequence flow data |
| CN114648245A (en) * | 2022-04-07 | 2022-06-21 | 南通俊朗智能科技有限公司 | Equipment inspection priority evaluation method and system based on artificial intelligence |
| CN116012579A (en) * | 2023-02-17 | 2023-04-25 | 厦门破云溯星科技有限公司 | Method for detecting abnormal states of parts based on photographed images of intelligent inspection robot of train |
| CN116665126A (en) * | 2023-05-29 | 2023-08-29 | 华中科技大学 | Robot inspection part defect detection method and application thereof |
| CN116630327A (en) * | 2023-07-25 | 2023-08-22 | 江苏太湖锅炉股份有限公司 | Boiler state abnormity monitoring system based on thermodynamic diagram |
| CN116702081A (en) * | 2023-08-07 | 2023-09-05 | 西安格蒂电力有限公司 | Intelligent inspection method for power distribution equipment based on artificial intelligence |
| CN116879662A (en) * | 2023-09-06 | 2023-10-13 | 山东华尚电气有限公司 | Transformer fault detection method based on data analysis |
Non-Patent Citations (3)
| Title |
|---|
| Process monitoring of abnormal working conditions in the zinc roasting process with an ALD-based LOF-PCA method;Zhenxiang Feng 等;《Process Safety and Environmental Protection》;第161卷;全文 * |
| 基于改进K-medoids聚类和SVM的异常用电模式在线检测方法;胡聪 等;《国外电子测量技术》;第41卷(第2期);全文 * |
| 基于自步学习的刀具加工过程监测数据异常检测方法;张建 等;《上海交通大学学报》;第57卷(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117349711A (en) | 2024-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117349711B (en) | Electronic tag data processing method and system for railway locomotive parts | |
| KR101948604B1 (en) | Method and device for equipment health monitoring based on sensor clustering | |
| CN111555716A (en) | Method, device and equipment for determining working state of photovoltaic array and storage medium | |
| CN112284440B (en) | Sensor data deviation self-adaptive correction method | |
| JP6871877B2 (en) | Information processing equipment, information processing methods and computer programs | |
| CN111538311B (en) | Flexible multi-state self-adaptive early warning method and device for mechanical equipment based on data mining | |
| CN113609901A (en) | Power transmission and transformation equipment fault monitoring method and system | |
| CN114997256A (en) | Method and device for detecting abnormal power of wind power plant and storage medium | |
| CN116108402A (en) | Method, equipment and storage medium based on electric power multi-source heterogeneous data fusion analysis | |
| CN114783165A (en) | Cable channel external damage prevention online monitoring system based on distributed optical fiber vibration sensing | |
| KR101960755B1 (en) | Method and apparatus of generating unacquired power data | |
| Huang et al. | A method of transmission conductor-loosened detect based on image sensors | |
| CN111314110B (en) | Fault early warning method for distributed system | |
| CN117331790A (en) | Machine room fault detection method and device for data center | |
| CN115495274B (en) | Exception handling method based on time sequence data, network equipment and readable storage medium | |
| CN117351271A (en) | Fault monitoring method and system for high-voltage distribution line monitoring equipment and storage medium thereof | |
| CN115797355B (en) | Fault detection method and system for solar photovoltaic power generation device | |
| CN117150298A (en) | Deep learning-based subway FAS fire alarm system debugging method | |
| KR102158100B1 (en) | Auto monitoring method and apparatus by using anomaly detection | |
| CN114492636B (en) | Transformer winding state signal acquisition system | |
| CN116401137A (en) | Core particle health state prediction method and device, electronic equipment and storage medium | |
| WO2020170304A1 (en) | Learning device and method, prediction device and method, and computer-readable medium | |
| CN115511106B (en) | Method, device and readable storage medium for generating training data based on time sequence data | |
| CN117783792B (en) | Valve side sleeve insulation state detection method and system based on multiparameter real-time monitoring | |
| CN113642618B (en) | Method and equipment for training screw device state prediction model |
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 | ||
| GR01 | Patent grant |