CN116975672B - Temperature monitoring method and system for coal mine belt conveying motor - Google Patents
Temperature monitoring method and system for coal mine belt conveying motor Download PDFInfo
- Publication number
- CN116975672B CN116975672B CN202311226299.8A CN202311226299A CN116975672B CN 116975672 B CN116975672 B CN 116975672B CN 202311226299 A CN202311226299 A CN 202311226299A CN 116975672 B CN116975672 B CN 116975672B
- Authority
- CN
- China
- Prior art keywords
- cluster
- data
- class
- membership
- temperature
- 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
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000012544 monitoring process Methods 0.000 title claims abstract description 27
- 239000003245 coal Substances 0.000 title claims abstract description 23
- 238000012216 screening Methods 0.000 claims abstract description 6
- 238000005259 measurement Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 2
- 230000008859 change Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Probability & Statistics with Applications (AREA)
- Control Of Conveyors (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a method and a system for monitoring the temperature of a coal mine belt conveying motor, comprising the following steps: collecting motor temperature data, determining a data sample according to the motor temperature data, clustering the data sample by using different K values, constructing a plurality of cluster chains according to the difference degree of the clusters corresponding to the different K values, determining the membership degree of each data point in the data sample to each cluster chain, screening stable cluster chains and the membership data points of the stable cluster chains according to the membership degree, determining an optimal K value according to the stable cluster chains, acquiring an optimal initial clustering center according to the distribution of the membership data points of each stable cluster chain, clustering the data sample according to the optimal K value and all the optimal initial clustering centers, and monitoring the motor temperature data according to the clustering result. The invention has good clustering effect on the motor temperature data and ensures the accuracy of motor temperature monitoring.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a system for monitoring the temperature of a coal mine belt conveying motor.
Background
In the working process of the coal mine belt motor, the heating efficiency of different devices inside the motor is different, the temperature distribution of different areas is different on the surface of the motor, the temperature is conductive, and the temperature area can not be divided directly according to the components inside the motor. The temperature of different areas of the motor can be clustered by adopting clustering, so that different temperature areas of the motor are obtained, and the abnormal temperature is measured according to the temperature areas.
When clustering is carried out on the motor surface temperature data, the working efficiency of different motors is different, and under the condition of data distribution in an undefined area, the clustering quality is poor due to the fixed K value and the initial clustering center.
Disclosure of Invention
In order to solve the problems, the invention provides a temperature monitoring method and a temperature monitoring system for a coal mine belt conveying motor.
The invention discloses a temperature monitoring method of a coal mine belt conveying motor, which adopts the following technical scheme:
the embodiment of the invention provides a temperature monitoring method for a coal mine belt conveying motor, which comprises the following steps:
collecting motor temperature data, and determining a data sample according to the motor temperature data;
presetting a plurality of different K values, and clustering the data samples by using the different K values to obtain a plurality of clusters corresponding to the different K values; constructing a plurality of class cluster chains according to the difference degrees of class clusters corresponding to different K values, wherein each class cluster chain consists of class clusters corresponding to different K values;
determining the membership degree of each data point in the data sample to each class cluster chain; screening the stable cluster chains and the membership data points of the stable cluster chains according to the membership degree, and determining the optimal according to the stable cluster chainsA value; acquiring an optimal initial clustering center according to the distribution of the membership data points of each stable cluster chain;
according to the bestAnd clustering the data samples by the values and all the optimal initial clustering centers, and monitoring the motor temperature data according to the clustering result.
Preferably, the constructing a plurality of cluster chains according to the degree of difference of the clusters corresponding to different K values includes the following specific steps:
for each class cluster, determining the center point of the class cluster; determining the difference degree of any two clusters according to the difference of the central points of the two clusters under different K values; for the firstEach class cluster at K value, comparing class cluster with +.>The difference degree between each class cluster under the K value is determined, and the +.>Cluster under K value as +.>Linking class clusters of class clusters under K values; and determining a plurality of class cluster chains according to the chained class clusters of each class cluster.
Preferably, the determining a plurality of cluster chains according to the linked cluster of each cluster comprises the following specific steps:
for each class cluster, if the class cluster is not a chained class cluster of any one class cluster, the class cluster is used as a starting class cluster;
and linking the linked class clusters of each class cluster after each class cluster to form a plurality of class cluster chains starting from each initial class cluster and ending from the class cluster corresponding to the maximum K value.
Preferably, the determining the membership degree of each data point in the data sample to each cluster chain comprises the following specific steps:
wherein,representing the%>Data point pair->Membership of the bar cluster chain; />Indicate->The number of class clusters contained in the strip class cluster chain; />Representing the%>Data points; />Indicate->The>A cluster of classes; />As a conditional function +.>Indicating +.>Data points belong to->The>When the cluster is of the class, return result 1, data sample +.>Data points do not belong to->The>And when the cluster is clustered, returning a result 0.
Preferably, the step of screening the stable cluster chain and the membership data points of the stable cluster chain according to the membership degree includes the following specific steps:
preset membership thresholdProportional threshold +.>When the membership degree of the data point to the class cluster chain is greater than or equal to a membership degree threshold, the data point is used as the membership data point of the class cluster chain;
for each cluster chain, acquiring a union set of data points contained in all clusters on the cluster chain, taking the number of the data points contained in the acquired union set as a first number, acquiring the ratio of the number of the membership data points of the cluster chain to the first number, and taking the ratio as a stable ratio of the cluster chain, wherein if the stable ratio of the cluster chain exceeds a proportional threshold valueCluster-like chains are denoted as stable cluster-like chains.
Preferably, the obtaining an optimal initial clustering center according to the distribution of the membership data points of each stable cluster chain includes the following specific steps:
for each stable cluster chain, dividing all the membership data points of the stable cluster chain into a plurality of data point categories according to the membership degree of the membership data points to the stable cluster chain; determining a center point for each data point class; determining the difference between each membership data point and each data point class according to the difference between each membership data point of the stable class cluster chain and the center point of each data point class; determining the optimal initial clustering center probability of each membership data point according to the difference between each membership data point and each data point category and the membership degree of all data points in each data point category; and taking the membership data point with the maximum probability of the optimal initial clustering center as the optimal initial clustering center.
Preferably, the determining the optimal initial cluster center probability of each membership data point includes the following specific steps:
wherein,to stabilize cluster chain +.>Optimal initial cluster center probability of each membership data point; />Representing the +.sub.under the stable cluster chain>The average of membership of all data points in a class of data points; />To stabilize cluster chain +.>Membership data points and->Variability in data point categories; />The number of data point categories under the stable class cluster chain; />Is an exponential function with a base of natural constant.
Preferably, the clustering is performed on the data samples by using different K values to obtain a plurality of clusters corresponding to the different K values, including the following specific steps:
constructing a distance measurement formula:
wherein,representing the%>Data points and->Distance of data points; />Representing the%>Data points and->Euclidean distance between the individual data points; />Representing the%>Data points and->Differences in temperature data between data points; />Is an absolute value symbol; />Representation normalization;
and clustering the data samples according to each K value by adopting a distance measurement formula to obtain a plurality of clusters corresponding to each K value.
Preferably, the monitoring the motor temperature data according to the clustering result includes the following specific steps:
for each cluster in the clustering result, a convex hull area of the cluster is obtained and used as a temperature area, and the maximum temperature data in the cluster is used as a temperature areaMinimum temperature data ∈ ->Acquiring a reference temperature range of the temperature region；
And acquiring temperature data of each position of the motor surface at each moment according to the position, and acquiring a temperature region in which the temperature data is positioned, wherein when the temperature data is out of a reference temperature range of the temperature region in which the temperature data is positioned, the temperature data of the position is abnormal.
The invention also provides a coal mine belt conveying motor temperature monitoring system, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of any one of the coal mine belt conveying motor temperature monitoring methods when executing the computer program.
The technical scheme of the invention has the beneficial effects that: according to the invention, through collecting motor temperature data, determining a data sample according to the motor temperature data, clustering the data sample by utilizing different K values, constructing a plurality of class cluster chains according to the difference degree of class clusters corresponding to the different K values, determining the membership degree of each data point in the data sample to each class cluster chain, screening the stable class cluster chains and the membership data points of the stable class cluster chains according to the membership degree, and determining the optimal according to the stable class cluster chainsThe value, according to the distribution of the membership data points of each stable cluster chain, the initial cluster center is obtained, and according to the different membership degrees, the data points with smaller difference with the higher membership degree are selected as the initial clustersThe center ensures that the subsequent clustering is carried out according to the initial clustering center, so that the membership data points contained in the class clusters of the stable class cluster chain can be better clustered into one class, and the clustering result is more accurate; the invention is based on the best->And clustering the data samples by the values and all initial clustering centers, and monitoring the motor temperature data according to the clustering result. The invention has good clustering effect on the motor temperature data and ensures the accuracy of motor temperature detection.
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 the steps of a method for monitoring the temperature of a coal mine belt conveyor motor.
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 a specific implementation, structure, characteristics and effects of the temperature monitoring method for the coal mine belt conveying motor according to the invention in combination with 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 coal mine belt conveying motor temperature monitoring method, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for monitoring temperature of a coal mine belt conveyor motor according to an embodiment of the invention is shown, the method includes the following steps:
s001, collecting motor temperature data.
And acquiring an infrared image of the coal mine belt conveying motor, and converting electromagnetic radiation energy values in the infrared image into corresponding temperatures through an infrared data converter to obtain temperature data of each position on the surface of the coal mine belt conveying motor.
Thus, temperature data for each position of the motor surface is obtained.
S002, clustering different K values of the motor temperature data to obtain clustering results under each K value.
Since the temperature is conductive, it is not possible to obtain an accurate temperature division region directly by the different constituent parts of the motor. Therefore, the embodiment of the invention obtains accurate temperature dividing areas by clustering the temperature data of the motor which normally works. In the clustering, if the K values adopted are different, the clustering results are different, and the optimal K value cannot be predicted because the conduction condition of the temperature of the surface of the motor is not known. According to the embodiment of the invention, the clustering result corresponding to each K value is analyzed by setting a plurality of K values, so that the optimal K value and the optimal initial clustering center are obtained, and the accurate temperature division area is obtained by clustering according to the optimal initial clustering center.
In the embodiment of the present invention, a plurality of K values are preset, and the embodiment of the present invention is described by taking k=3, 4, 5, 6, 7, 8, and 9 as examples, which are not limited in particular, and the implementation personnel can set the K values according to specific implementation situations.
When the motor works normally, temperature data of all positions on the surface of the motor are used as data samples, and a distance measurement formula is built according to the position difference of two data points in the data samples and the difference of the temperature data:
wherein,representing the%>Data points and->Distance of data points; />Representing the%>Data points and->Euclidean distance between the individual data points; />Representing the%>Data points and->Differences in temperature data between data points; />Is an absolute value symbol; due to European distance->Absolute value of difference from temperature data +.>There is an order of magnitude difference, thus for Euclidean distance respectively +>Absolute value of difference from temperature data +.>Normalizing to obtain normalized result->、/>，/>Representation normalization; when normalized Euclidean distance->Smaller, at the same time, absolute value of difference of normalized temperature data +.>Also smaller, the +.>Data points and->The smaller the distance of the data points. It should be noted that the embodiment of the present invention is directed to Euclidean distance +.>Absolute value of difference from temperature data +.>In the case of normalizing, the normalization by the Z-Score is described as an example, but the method is not particularly limited, and in other embodiments, the practitioner may select the normalization method according to the actual implementation.
And clustering the data samples according to each K value by adopting a distance measurement formula to obtain a clustering result corresponding to each K value. In the embodiment of the invention, K-means clustering is taken as an example for explanation, the clustering method is not limited, and in other embodiments, an implementation person can set the clustering method according to specific implementation conditions.
So far, the clustering result corresponding to each K value is obtained.
S003, carrying out differential degree measurement on class clusters corresponding to different K values, and obtaining a plurality of class cluster chains.
It should be noted that, data points in class clusters with different K values may be basically consistent, and at this time, the corresponding class cluster has smaller change under different K values, so the embodiment of the invention analyzes differences of class clusters with different K values, constructs a class cluster chain according to the class clusters with different K values with small differences, so as to obtain the best according to the class cluster chain subsequentlyValues.
In the embodiment of the invention, for each class cluster in the clustering result of each K value, the distance between each data point in the class cluster and each other data point in the class cluster is obtained by using a distance measurement formula, and the sum of the distances between each data point and each other data point in the class cluster is used as the first difference of each data point. And taking the data point with the smallest first difference in the class cluster as the center point of the class cluster.
And obtaining the distance between the center points of any two clusters under different K values by using a distance measurement formula, and taking the distance as the difference degree of the two clusters. Acquisition of the firstEach cluster under K value is associated with +.>The degree of difference between each cluster of classes at the value of K,for->Every cluster under K value, will be +.>And the class cluster with the smallest difference degree with the class cluster under the K value is used as a linking class cluster of the class cluster.
For the firstFor each cluster of classes at the value of K, and (2)>If the cluster is not a chained cluster of any cluster, the cluster is used as a starting cluster. All the initial class clusters are acquired.
And linking the linked class clusters of each class cluster after each class cluster to form a plurality of class cluster chains starting from each initial class cluster and ending from the class cluster corresponding to the maximum K value.
Thus, a plurality of cluster-like chains are obtained.
S004, obtaining the membership degree of motor temperature data to each cluster chain, and obtaining a plurality of stable cluster chains and the best according to the membership degreeValues.
It should be noted that, if the data points all belong to the same cluster in the same cluster chain under different K value clusters, it is indicated that the data points have a large membership to the cluster chain, and the probability that all the data points having a large membership to the cluster chain belong to the same part of the temperature region of the motor is larger. Therefore, according to the class clusters to which the data points belong under different K values, the embodiment of the invention calculates the membership degree of each data point to each class cluster chain so as to screen the class cluster chains according to the membership degree later and obtain the class cluster chains corresponding to the temperature area of the same part of the motor.
In the embodiment of the invention, for each data point in a data sample, the membership degree of the data point to each cluster chain of the class is calculated:
wherein,representing the%>Data pointsFor->Membership of the bar cluster chain; />Indicate->The number of class clusters contained in the strip class cluster chain; />Representing the%>Data points; />Indicate->The>A cluster of classes; />As a conditional function +.>Indicating +.>Data points belong to->The>When the cluster is of the class, return result 1, data sample +.>Data points do not belong to->The>When the cluster is clustered, returning a result 0; />For counting the +.>Data points belong to->The number of class clusters in the strip class cluster chain is +.>The ratio of all clusters in the chain of cluster types, when the ratio is larger, the +.>Data point pair->The greater the membership of the bar-type cluster chain.
Preset membership thresholdProportional threshold +.>Embodiments of the invention are described in->，/>For example, without limitation, the practitioner may set the membership threshold and the proportion threshold according to the specific implementation.
When the membership degree of the data point to the class cluster chain is greater than or equal to the membership degree threshold, the data point is used as the membership data point of the class cluster chain. All the membership data points of each cluster chain are acquired.
For each cluster chain, acquiring a union set of data points contained in all clusters on the cluster chain, taking the number of the data points contained in the acquired union set as a first number, acquiring the ratio of the number of the membership data points of the cluster chain to the first number, and taking the ratio as a stable ratio of the cluster chain, wherein if the stable ratio of the cluster chain exceeds a proportional threshold valueAnd considering that all class clusters contained in the class cluster chain are stable under different K values, and marking the class cluster chain as a stable class cluster chain.
Thus, a plurality of stable cluster chains are obtained.
It should be noted that the change of the class clusters in the stable class cluster chain under different K values is smaller, so that all data points with high membership in the class clusters in the stable class cluster chain are more likely to belong to the same temperature region on the motor. Therefore, each stable cluster chain corresponds to a temperature region in the following process, and therefore the number of the stable cluster chains is the number of the temperature regions. In order to ensure that the data points in each temperature region can be clustered into the same class cluster in the subsequent clustering, the set K value needs to be the same as the number of the temperature regions, namely the same as the number of stable class cluster chains.
In the embodiment of the invention, the number of stable cluster chains is taken as the optimal numberValues.
Thus, a stable cluster chain and the best are obtainedValues.
S005, classifying the motor temperature data according to the membership degree to obtain an optimal initial clustering center.
It should be noted that, the change of the class clusters in the stable class cluster chain under different K values is smaller, so that the more similar the temperature data in the region where the data points in the class clusters corresponding to the stable class cluster chain are located, the more likely the data points in the class clusters corresponding to the stable class cluster chain are the temperature regions of the same part on the motor. If the membership degree of the membership data points of the stable cluster chain is different, if the centers of all elements with high membership degree are only adopted as the initial clustering centers of the region, the clustering effect on the membership data points with lower membership degree is poor, and in order to obtain the optimal initial clustering center of the region corresponding to the stable cluster chain, the membership data points are firstly classified according to the membership degree, so that the optimal initial clustering center is obtained according to the membership data points of different categories.
In the embodiment of the invention, for each stable cluster chain, all the membership data points of the stable cluster chain are divided intoData point class, embodiments of the invention are described in +.>Examples are described without limitation. Respectively subject the membership degree to、/>、…、/>All membership data points within the range act as one class of data points. For example when->，/>When the membership degree is +.>The membership data points in the range are a data point class, and the membership degree is +.>The membership data points in the range are a data point class, and the membership degree is +.>Membership data points within the range are a class of data points.
For each data point class under each stable class cluster chain, using a distance measurement formula to obtain the distance from each data point in the data point class to each other data point in the data point class, and taking the sum of the distances of each data point and each other data point in the data point class as a second difference of each data point. The data point with the smallest second difference in the data point class is taken as the center point of the data point class.
It should be noted that, the center point of the data point class can reflect the average data size distribution of the data points in the data point class, in order to completely cluster the membership data points under different membership degrees, the optimal initial clustering center needs to be similar to the center point of each data point class, and the optimal initial clustering center should be more similar to the data point class with larger membership degrees.
In the embodiment of the invention, for each membership data point of each stable class cluster chain, a distance from the membership data point to the center point of each data point class under the stable class cluster is calculated by using a distance measurement formula and used as the difference between the membership data point and each data point class. And obtaining the average value of the membership degree of all data points in each data point category under the stable category cluster.
According to the average value of membership degrees of all data points in each data point category under the stable category cluster and the difference between the membership data points and each data point category, acquiring initial cluster center probability of the membership data points:
wherein,to stabilize cluster chain +.>Each membership data pointInitial cluster center probability; />Representing the +.sub.under the stable cluster chain>The average of membership of all data points in a class of data points; />To stabilize cluster chain +.>Membership data points and->Variability in data point categories; />The number of data point categories under the stable class cluster chain; />Is an exponential function with a natural constant as a base; embodiments of the invention will->As a stable cluster under the chain +.>Weight of data point class, when +.>The larger the more interesting the +.>Membership data points and->Differences in data point categories, when +.>Membership data points and pointsThe smaller the difference in the categories of the data points is, the +.>The better the clustering effect of the individual membership data points as initial cluster center, at this point +.>The greater the initial cluster center probability of each membership data point.
And taking the membership data point with the highest probability of the initial clustering center as the optimal initial clustering center. And acquiring all the optimal initial cluster centers.
So far, the optimal initial clustering center is obtained.
S006 according to the bestClustering the motor temperature data by the value and the optimal initial clustering center, and carrying out anomaly analysis on the motor temperature data at each moment according to the clustering result.
By using the bestAnd clustering the data samples according to the optimal initial clustering center to obtain a final clustering result, wherein the final clustering result reflects the temperature distribution of different components of the motor under the normal operation of the motor. For each class cluster in the final clustering result, acquiring a convex hull area of the class cluster as a temperature area, and taking the maximum temperature data in the class clusterMinimum temperature data ∈ ->Temperature maximum boundary +.>And temperature minimum boundary->Will be/>As a reference temperature range for the temperature region.
And acquiring temperature data of each position of the motor surface at each moment according to the position, wherein the temperature data of the position is normal when the temperature data is within a reference temperature range of the temperature region to which the temperature data belongs, and otherwise, the temperature data of the position is abnormal when the temperature data is outside the reference temperature range of the temperature region to which the temperature data belongs.
When abnormal temperature data exist, alarming is carried out, and related personnel are reminded of overhauling the position corresponding to the abnormal temperature data in the motor.
The embodiment of the invention also provides a temperature monitoring system of the coal mine belt conveying motor, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of any one of the temperature monitoring methods of the coal mine belt conveying motor when executing the computer program.
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 (5)
1. The method for monitoring the temperature of the coal mine belt conveying motor is characterized by comprising the following steps of:
collecting motor temperature data, and determining a data sample according to the motor temperature data;
presetting a plurality of different K values, and clustering the data samples by using the different K values to obtain a plurality of clusters corresponding to the different K values; constructing a plurality of class cluster chains according to the difference degrees of class clusters corresponding to different K values, wherein each class cluster chain consists of class clusters corresponding to different K values;
determining the membership degree of each data point in the data sample to each class cluster chain; screening stable cluster chains according to the membership degreeAnd the membership data points of the stable cluster chain, determining the best according to the stable cluster chainA value; acquiring an optimal initial clustering center according to the distribution of the membership data points of each stable cluster chain;
according to the bestClustering the data samples by the values and all the optimal initial clustering centers, and monitoring the motor temperature data according to the clustering result;
according to the difference degree of class clusters corresponding to different K values, constructing a plurality of class cluster chains, wherein the specific steps are as follows:
for each class cluster, determining the center point of the class cluster; determining the difference degree of any two clusters according to the difference of the central points of the two clusters under different K values; for the firstEach class cluster at K value, comparing class cluster with +.>The difference degree between each class cluster under the K value is determined, and the +.>Cluster under K value as +.>Linking class clusters of class clusters under K values; determining a plurality of class cluster chains according to the chained class clusters of each class cluster;
the method comprises the following specific steps of determining a plurality of class cluster chains according to the chained class clusters of each class cluster:
for each class cluster, if the class cluster is not a chained class cluster of any one class cluster, the class cluster is used as a starting class cluster;
linking the linked class clusters of each class cluster after each class cluster to form a plurality of class cluster chains starting from each initial class cluster and ending from the class cluster corresponding to the maximum K value;
the determining the membership degree of each data point in the data sample to each cluster chain comprises the following specific steps:
wherein,representing the%>Data point pair->Membership of the bar cluster chain; />Indicate->The number of class clusters contained in the strip class cluster chain; />Representing the%>Data points; />Indicate->The>A cluster of classes; />As a conditional function +.>Indicating +.>Data points belong to->The>When the cluster is of the class, return result 1, data sample +.>Data points do not belong to->The>When the cluster is clustered, returning a result 0;
the clustering is carried out on the data samples by using different K values to obtain a plurality of clusters corresponding to the different K values, and the method comprises the following specific steps:
constructing a distance measurement formula:
wherein,representing the%>Data points and->Distance of data points; />Representing the%>Data points and->Euclidean distance between the individual data points; />Representing the%>Data points and->Differences in temperature data between data points; />Is an absolute value symbol; />Representation normalization;
clustering the data samples according to each K value by adopting a distance measurement formula to obtain a plurality of clusters corresponding to each K value;
the motor temperature data is monitored according to the clustering result, and the method comprises the following specific steps:
for each cluster in the clustering result, a convex hull area of the cluster is obtained and used as a temperature area, and the maximum temperature data in the cluster is used as a temperature areaMinimum temperature data ∈ ->Acquiring a reference temperature range of the temperature region；
And acquiring temperature data of each position of the motor surface at each moment according to the position, and acquiring a temperature region in which the temperature data is positioned, wherein when the temperature data is out of a reference temperature range of the temperature region in which the temperature data is positioned, the temperature data of the position is abnormal.
2. The method for monitoring the temperature of the coal mine belt conveyor motor according to claim 1, wherein the step of screening the stable cluster chains and the membership data points of the stable cluster chains according to the membership degree comprises the following specific steps:
preset membership thresholdProportional threshold +.>When the membership degree of the data point to the class cluster chain is greater than or equal to a membership degree threshold, the data point is used as the membership data point of the class cluster chain;
for each cluster chain, acquiring a union set of data points contained in all clusters on the cluster chain, taking the number of the data points contained in the acquired union set as a first number, acquiring the ratio of the number of the membership data points of the cluster chain to the first number, and taking the ratio as a stable ratio of the cluster chain, wherein if the stable ratio of the cluster chain exceeds a proportional threshold valueCluster-like chains are denoted as stable cluster-like chains.
3. The method for monitoring the temperature of the coal mine belt conveying motor according to claim 1, wherein the obtaining of the optimal initial clustering center according to the distribution of the membership data points of each stable cluster chain comprises the following specific steps:
for each stable cluster chain, dividing all the membership data points of the stable cluster chain into a plurality of data point categories according to the membership degree of the membership data points to the stable cluster chain; determining a center point for each data point class; determining the difference between each membership data point and each data point class according to the difference between each membership data point of the stable class cluster chain and the center point of each data point class; determining the optimal initial clustering center probability of each membership data point according to the difference between each membership data point and each data point category and the membership degree of all data points in each data point category; and taking the membership data point with the maximum probability of the optimal initial clustering center as the optimal initial clustering center.
4. A method for monitoring the temperature of a coal mine belt conveyor motor according to claim 3, wherein the determining of the optimal initial cluster center probability for each membership data point comprises the following specific steps:
wherein,to stabilize cluster chain +.>Optimal initial cluster center probability of each membership data point; />Representing the +.sub.under the stable cluster chain>The average of membership of all data points in a class of data points; />To stabilize cluster chain +.>Membership data points and->Variability in data point categories; />The number of data point categories under the stable class cluster chain; />Is an exponential function with a base of natural constant.
5. A coal mine belt conveyor motor temperature monitoring system comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor, when executing the computer program, carries out the steps of the method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311226299.8A CN116975672B (en) | 2023-09-22 | 2023-09-22 | Temperature monitoring method and system for coal mine belt conveying motor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311226299.8A CN116975672B (en) | 2023-09-22 | 2023-09-22 | Temperature monitoring method and system for coal mine belt conveying motor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116975672A CN116975672A (en) | 2023-10-31 |
| CN116975672B true CN116975672B (en) | 2023-12-15 |
Family
ID=88478206
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311226299.8A Active CN116975672B (en) | 2023-09-22 | 2023-09-22 | Temperature monitoring method and system for coal mine belt conveying motor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116975672B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117421620B (en) * | 2023-12-18 | 2024-02-27 | 北京云摩科技股份有限公司 | Interaction method of tension state data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858544A (en) * | 2019-01-28 | 2019-06-07 | 重庆邮电大学 | The steel product quality detection method clustered based on section shade collection and density peaks |
| WO2019218263A1 (en) * | 2018-05-16 | 2019-11-21 | 深圳大学 | Extreme learning machine-based extreme ts fuzzy inference method and system |
| CN115017988A (en) * | 2022-06-01 | 2022-09-06 | 东南大学 | Competitive clustering method for state anomaly diagnosis |
-
2023
- 2023-09-22 CN CN202311226299.8A patent/CN116975672B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019218263A1 (en) * | 2018-05-16 | 2019-11-21 | 深圳大学 | Extreme learning machine-based extreme ts fuzzy inference method and system |
| CN109858544A (en) * | 2019-01-28 | 2019-06-07 | 重庆邮电大学 | The steel product quality detection method clustered based on section shade collection and density peaks |
| CN115017988A (en) * | 2022-06-01 | 2022-09-06 | 东南大学 | Competitive clustering method for state anomaly diagnosis |
Non-Patent Citations (2)
| Title |
|---|
| Sea surface temperature clustering based on type-2 fuzzy theory;Kun Qin等;《 IEEE Xplore》;全文 * |
| 基于并行模糊C-均值聚类的风电机组发电机故障诊断研究;孙鹤旭;孙泽贤;张靖轩;;太阳能学报(03);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116975672A (en) | 2023-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116975672B (en) | Temperature monitoring method and system for coal mine belt conveying motor | |
| CN117349220B (en) | Data processing method and system based on PCI bus | |
| CN109726649B (en) | Remote sensing image cloud detection method and system and electronic equipment | |
| CN115994907B (en) | Intelligent processing system and method for comprehensive information of food detection mechanism | |
| CN110717540A (en) | Method and device for identifying new radar source individuals | |
| WO2018006631A1 (en) | User level automatic segmentation method and system | |
| CN112465001B (en) | Classification method and device based on logistic regression | |
| CN115876258A (en) | Livestock and poultry breeding environment abnormity monitoring and alarming system based on multi-source data | |
| CN117473351B (en) | Power supply information remote transmission system based on Internet of things | |
| CN117789038B (en) | Training method of data processing and recognition model based on machine learning | |
| CN117173640A (en) | Beef cattle temperature anomaly detection method | |
| CN113204990B (en) | Machine learning method and device based on intelligent fishing rod | |
| CN111027841A (en) | Low-voltage transformer area line loss calculation method based on gradient lifting decision tree | |
| CN110889435A (en) | Insulator evaluation classification method and device based on infrared image | |
| CN117170979B (en) | Energy consumption data processing method, system, equipment and medium for large-scale equipment | |
| CN115841491B (en) | Quality detection method for porous metal material | |
| CN116205923B (en) | Nondestructive testing method for internal defects of automobile hub based on X-RAY | |
| CN107895361A (en) | A kind of hyperspectral abnormity detection method based on local density's purifying background | |
| CN116206208A (en) | Forestry plant diseases and insect pests rapid analysis system based on artificial intelligence | |
| CN108415958B (en) | Weight processing method and device for index weight VLAD features | |
| CN116484206A (en) | SEIM-based unknown radiation source individual identification method and system | |
| CN116229330A (en) | Method, system, electronic equipment and storage medium for determining video effective frames | |
| CN113808088A (en) | Pollution detection method and system | |
| KR100581673B1 (en) | Data Classification Method | |
| CN108537092B (en) | Variant red blood cell identification method and device |
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 |