CN117076258A - Remote monitoring method and system based on Internet cloud - Google Patents
Remote monitoring method and system based on Internet cloud Download PDFInfo
- Publication number
- CN117076258A CN117076258A CN202311328782.7A CN202311328782A CN117076258A CN 117076258 A CN117076258 A CN 117076258A CN 202311328782 A CN202311328782 A CN 202311328782A CN 117076258 A CN117076258 A CN 117076258A
- Authority
- CN
- China
- Prior art keywords
- equipment
- data
- cloud
- feature
- remote monitoring
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000012544 monitoring process Methods 0.000 title claims abstract description 37
- 230000002159 abnormal effect Effects 0.000 claims abstract description 42
- 238000004458 analytical method Methods 0.000 claims abstract description 22
- 238000012216 screening Methods 0.000 claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 238000012300 Sequence Analysis Methods 0.000 claims abstract description 10
- 230000005856 abnormality Effects 0.000 claims abstract description 10
- 238000007621 cluster analysis Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 18
- 230000000737 periodic effect Effects 0.000 claims description 17
- 238000003860 storage Methods 0.000 claims description 12
- 238000003064 k means clustering Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 206010000117 Abnormal behaviour Diseases 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 230000000875 corresponding effect Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000013450 outlier detection Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 4
- 238000012731 temporal analysis Methods 0.000 claims description 4
- 238000000700 time series analysis Methods 0.000 claims description 4
- 230000002596 correlated effect Effects 0.000 claims description 3
- 238000005304 joining Methods 0.000 claims description 3
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a remote monitoring method and a remote monitoring system based on an Internet cloud, which relate to the technical field of monitoring, and the remote monitoring method based on the Internet cloud comprises the following steps: acquiring equipment operation parameters; screening out important features with the greatest influence on equipment abnormality early warning judgment; calculating a weight value of the importance feature by using the information gain; marking the equipment area with abnormal change as a potential problem area; and constructing a device fault prediction model in the cloud by using a beamforming method and a time sequence analysis method. According to the invention, through feature importance analysis, information gain calculation and equipment cluster analysis, abnormal equipment and potential problem areas can be accurately identified, so that not only is the accuracy of fault detection improved, but also equipment which is likely to be in fault at the next moment can be predicted, further, the maintenance and management of the equipment can be carried out in advance, and the efficiency of a remote monitoring system is improved.
Description
Technical Field
The invention relates to the technical field of monitoring, in particular to a remote monitoring method and system based on an Internet cloud.
Background
With the development of electronic information technology, network control systems have also been developed rapidly. The advent of embedded computer systems and embedded operating systems for such microelectronic products has provided the basis for the intellectualization and miniaturization of network control nodes. The advent of various network technologies has also provided different forms for the implementation of network control systems. The wireless communication technology reduces the space restriction factors of the network control system, the development of the information encryption technology improves the reliability and confidentiality of the network control system, and the reliability and stability of the complex control system designed by utilizing the component technology are greatly enhanced. In the modern informatization, network control is applied in various fields, such as: manufacturing, metallurgy, medical, transportation, and electrical, among others. In the field of people's social life, the application of network control will also be popularized, which shortens the distance between people and various social resources and changes the life style of people. With the development of globalization and internationalization trend, the application field of network control will be wider and wider. At the same time, the distance of remote control is also from several kilometers to several thousands of kilometers. The quality of the network control is also higher and higher, and the network control applied to the fields of medical treatment, military and the like has higher stability, real-time performance, reliability, safety and confidentiality.
With the development of the mobile internet, people have more demands for remotely grasping the current state of a target in real time through mobile terminals such as mobile phones. For example, technicians realize real-time monitoring of agricultural production processes through network connection of mobile phones; for another example, the vehicle owner can conveniently acquire real-time video information of the vehicle in a parking scene through mobile phone application software; for another example, a technician can remotely monitor the hardware equipment through a mobile phone or a computer, and can timely take corresponding measures to avoid potential safety hazards by monitoring whether the analysis equipment has abnormal conditions.
However, when the existing remote monitoring system monitors the running process of hardware equipment, the existing problems during the running of the equipment are inconvenient to find in time, early warning and corresponding measures cannot be taken for abnormal conditions in time, and failure prediction cannot be carried out on the equipment, so that the efficiency of the remote monitoring system is reduced, and particularly under the influence of noise or other interference factors, the potential abnormal conditions can be ignored, and the damage rate of the equipment is improved.
For the problems in the related art, no effective solution has been proposed at present.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a remote monitoring method and a remote monitoring system based on an Internet cloud, which solve the problems that the prior remote monitoring system is inconvenient to find out the problems existing in the running process of hardware equipment in time, early warning and corresponding measures are inconvenient to take in time, failure prediction can not be carried out on the equipment, the efficiency of the remote monitoring system is reduced, and particularly, under the influence of noise or other interference factors, the problems that some potential abnormal conditions are ignored and the damage rate of the equipment is improved are solved.
In order to achieve the above purpose, the invention is realized by the following technical scheme:
according to one aspect of the present invention, there is provided an internet cloud-based remote monitoring method, including the steps of:
s1, setting a plurality of sensors on equipment to be monitored, acquiring equipment operation parameters, and preprocessing;
s2, extracting features from the preprocessing result, and uploading the extracted features to a cloud for storage through the Internet;
s3, analyzing the feature importance of the extracted feature data, and screening out the importance feature with the greatest influence on the equipment abnormality early warning judgment;
s4, calculating a weight value of the importance feature by utilizing the information gain;
s5, carrying out cluster analysis on the equipment based on the weight value, and marking the equipment area with abnormal change as a potential problem area;
s6, constructing an equipment fault prediction model in the cloud by utilizing a beam forming method and a time sequence analysis method based on the potential problem area;
and S7, predicting the occurrence of the equipment fault at the next moment by using the equipment fault prediction model and the weight value in the cloud.
Further, setting a plurality of sensors on the equipment to be monitored, acquiring sensor data, and preprocessing the sensor data, wherein the steps comprise:
s11, collecting repeated data, missing values and abnormal values of equipment operation parameters acquired by each sensor, and denoising, filtering and smoothing the repeated data, the missing values and the abnormal values;
s12, connecting unprocessed data rows in the equipment operation parameters to generate a new data table, associating different data tables through external key values to generate a complete data table, and obtaining an accurate data set;
s13, determining external key relations among different data sets, connecting data in different data tables with each other according to requirements, creating a new data table, and associating through specified external key values;
s14, connecting the data tables to be joined together through the JOIN operators in the SQL sentences, and ensuring that the integrity constraint of the data is satisfied when the joining is carried out;
and S15, after the connection is completed, inserting test data to check whether the connection result is correct, so as to ensure that the connection result can be correctly identified and correlated, and obtaining accurate data of the operation parameters of the equipment.
Further, extracting features from the preprocessing result, and uploading the extracted features to a cloud for storage through the internet, wherein the method comprises the following steps:
s21, fusing accurate data of equipment operation parameters into the same data set by using a principal component analysis method;
s22, extracting relevant characteristics from the fused data set to obtain equipment operation parameter characteristics;
s23, uploading the obtained equipment operation parameter characteristics to a cloud end through the Internet for storage.
Further, performing feature importance analysis on the extracted feature data, and screening out importance features with the greatest influence on equipment abnormality early warning judgment, wherein the method comprises the following steps of:
s31, selecting and judging a characteristic evaluation index of the equipment abnormal behavior, and calculating the association degree between the equipment operation parameter characteristic and the equipment abnormal behavior based on the selected characteristic evaluation index;
s32, sorting the equipment operation parameter features based on the calculated association degree, wherein the higher the association degree is, the more important the equipment operation parameter features are, and the equipment operation parameter features with the highest association degree are taken as importance features.
Further, calculating the weight value of the importance feature using the information gain includes the steps of:
s41, collecting a sample data set of equipment operation parameters;
s42, calculating information entropy according to target variables in the sample data set, and measuring uncertainty of the target variables;
s43, for each feature, calculating the information gain between the feature and the target variable;
s44, taking the information gain value as a characteristic weight value, and carrying out normalization processing on the weight value;
s45, analyzing the normalized weight value, and knowing the importance degree of the operating parameter characteristics of the equipment.
Further, performing cluster analysis on the equipment based on the weight value, and marking the equipment area with abnormal change as a potential problem area comprises the following steps:
s51, taking the feature quantity of the importance features as the quantity of clustering centers, clustering the equipment areas by utilizing an improved K-Means clustering algorithm, and dividing the equipment areas into a plurality of groups;
s52, extracting data points from a plurality of groups, calculating local outlier factors of the data points by utilizing a local outlier detection algorithm, and judging whether the data points are outliers;
s53, counting the number of outliers in each group, and calculating the proportion of the outliers;
s54, if the proportion of the outliers in the group exceeds a preset threshold, judging that the group has abnormal change, and marking the equipment area with abnormal change as a potential problem area.
Further, the improved K-Means clustering algorithm has a calculation formula as follows:
;
wherein,nrepresenting the number of data points in the cluster;
da feature quantity expressed as an importance feature;
V{S,Hdenoted as noise measure;
Sirepresent the firstiClustering;
X ih expressed in clustersS i Middle (f)hFeature vectors of data points;
X jh representing all data points in the whole data set at the firsthAverage over individual features;
Hrepresented as a threshold;
jrepresenting a data point in the data.
Further, based on the potential problem area, constructing the equipment fault prediction model in the cloud by using a beamforming method and a time sequence analysis method comprises the following steps:
s61, analyzing whether trend items with longer periods exist in the operation parameters of the equipment by adopting a polynomial regression model;
s62, if the trend item exists, removing the trend item with a longer period;
s63, if the periodic variation exists in the operation parameters of the equipment, continuously analyzing the periodic variation;
s64, performing a beam forming method on the equipment operation parameters after removing the trend item according to the analysis result to obtain the amplitude and the phase of each frequency component;
s65, judging whether each frequency component is obvious or not by using saliency test, and extracting a salient period term to construct a period term model;
s66, regarding residual errors after eliminating trend items and period items as random changes, and constructing a residual error prediction model;
and S67, superposing the polynomial regression model, the periodic term model and the residual prediction model to obtain the equipment fault prediction model.
Further, predicting the occurrence of the equipment fault at the next moment by using the equipment fault prediction model and the weight value in the cloud comprises the following steps:
s71, predicting equipment operation parameters at the next moment through an equipment failure prediction model;
s72, carrying out corresponding weighting processing on the predicted equipment operation parameters and the weight values to obtain comprehensive prediction output of equipment failure occurrence.
According to another aspect of the present invention, there is also provided an internet cloud-based remote monitoring system, including:
the data acquisition and pretreatment module is used for setting a plurality of sensors on equipment to be monitored, acquiring equipment operation parameters and carrying out pretreatment;
the feature extraction module is used for extracting features from the preprocessing result and uploading the extracted features to the cloud for storage through the Internet;
the feature analysis module is used for carrying out feature importance analysis on the extracted feature data and screening out importance features with the greatest influence on equipment abnormality early warning judgment;
the weight calculation module is used for calculating the weight value of the importance characteristic by utilizing the information gain;
the abnormal change screening module is used for carrying out cluster analysis on the equipment based on the weight value and marking the equipment area with abnormal change as a potential problem area;
the equipment fault prediction model construction module is used for constructing an equipment fault prediction model in the cloud by utilizing a beam forming method and a time sequence analysis method based on the potential problem area;
the equipment fault prediction module is used for predicting equipment fault occurrence at the next moment by using the equipment fault prediction model and the weight value in the cloud;
the device comprises a preprocessing module, a weight calculation module, an abnormal change screening module, an equipment fault prediction model construction module, a device fault prediction model, a characteristic analysis module, a weight calculation module, an equipment fault prediction model, a weight calculation module and an abnormal change screening module.
The beneficial effects of the invention are as follows:
1. according to the invention, through feature importance analysis, information gain calculation and equipment cluster analysis, abnormal equipment and potential problem areas can be accurately identified, so that not only is the accuracy of fault detection improved, but also equipment fault prediction models can be constructed at the cloud end to predict equipment which is likely to be in fault at the next moment, further maintenance and management of the equipment can be performed in advance, sudden faults of the equipment are avoided, and the efficiency of a remote monitoring system is improved.
2. According to the invention, the device fault prediction model is constructed in the cloud by utilizing the beam forming method and the time sequence analysis method, so that the running trend of the device can be dynamically monitored and predicted, the tiny change of the performance of the device can be timely found, the device can be prevented and optimized before the fault occurs, the parameters of the prediction model can be dynamically adjusted according to the running state and environmental factors of the device, the intelligent decision is realized, the local resources can be greatly saved, the system performance and efficiency are improved, the running efficiency and reliability of the device are improved, and the running and maintenance cost of the device is reduced.
3. According to the invention, through an improved K-Means clustering algorithm, abnormal equipment can be effectively distinguished and marked as a potential problem area, so that equipment possibly having problems can be rapidly and effectively identified, and the equipment can be processed in time, thereby improving the efficiency of equipment management and maintenance and reducing equipment faults.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and 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 a remote monitoring method based on an Internet cloud according to an embodiment of the invention;
fig. 2 is a schematic block diagram of a remote monitoring system based on an internet cloud according to an embodiment of the present invention.
In the figure:
1. acquiring data and a preprocessing module; 2. a feature extraction module; 3. a feature analysis module; 4. a weight calculation module; 5. an abnormal change screening module; 6. the equipment fault prediction model building module; 7. and a device failure prediction module.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
In the description of the present invention, unless otherwise indicated, the meaning of "a plurality" is two or more. Furthermore, the terms "first," "second," "third," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
According to the embodiment of the invention, a remote monitoring method and a system based on an Internet cloud are provided.
Referring to the drawings and the specific embodiments, as shown in fig. 1, the remote monitoring method based on the internet cloud according to the embodiment of the invention includes the following steps:
s1, setting a plurality of sensors on equipment to be monitored, acquiring equipment operation parameters, and preprocessing;
specifically, the operating parameters of the device include temperature parameters, pressure parameters, electrical parameters, mechanical parameters, flow parameters, operating state parameters, environmental parameters, and the like.
S2, extracting features from the preprocessing result, and uploading the extracted features to a cloud for storage through the Internet;
s3, analyzing the feature importance of the extracted feature data, and screening out the importance feature with the greatest influence on the equipment abnormality early warning judgment;
s4, calculating a weight value of the importance feature by utilizing the information gain;
s5, carrying out cluster analysis on the equipment based on the weight value, and marking the equipment area with abnormal change as a potential problem area;
s6, constructing an equipment fault prediction model in the cloud by utilizing a beam forming method and a time sequence analysis method based on the potential problem area;
and S7, predicting the occurrence of the equipment fault at the next moment by using the equipment fault prediction model and the weight value in the cloud.
In one embodiment, setting a plurality of sensors on the device to be monitored, acquiring sensor data, and preprocessing includes the steps of:
s11, collecting repeated data, missing values and abnormal values of equipment operation parameters acquired by each sensor, and denoising, filtering and smoothing the repeated data, the missing values and the abnormal values;
s12, connecting unprocessed data rows in the equipment operation parameters to generate a new data table, associating different data tables through external key values to generate a complete data table, and obtaining an accurate data set;
s13, determining external key relations among different data sets, connecting data in different data tables with each other according to requirements, creating a new data table, and associating through specified external key values;
s14, connecting the data tables to be joined together through the JOIN operators in the SQL sentences, and ensuring that the integrity constraint of the data is satisfied when the joining is carried out;
and S15, after the connection is completed, inserting test data to check whether the connection result is correct, so as to ensure that the connection result can be correctly identified and correlated, and obtaining accurate data of the operation parameters of the equipment.
In one embodiment, extracting features from the preprocessing result, and uploading the extracted features to a cloud for storage through the internet includes the following steps:
s21, fusing accurate data of equipment operation parameters into the same data set by using a principal component analysis method;
s22, extracting relevant characteristics from the fused data set to obtain equipment operation parameter characteristics;
s23, uploading the obtained equipment operation parameter characteristics to a cloud end through the Internet for storage.
In one embodiment, performing feature importance analysis on the extracted feature data, and screening out importance features with the greatest influence on equipment abnormality early warning judgment comprises the following steps:
s31, selecting and judging a characteristic evaluation index of the equipment abnormal behavior, and calculating the association degree between the equipment operation parameter characteristic and the equipment abnormal behavior based on the selected characteristic evaluation index;
specifically, the abnormal behavior of the device includes that the device parameter exceeds the normal range, the input and output signals of the device are distorted, the working state of the device is unstable, the device cannot normally respond to the operation or the instruction, the device makes abnormal sound, and the like.
S32, sorting the equipment operation parameter features based on the calculated association degree, wherein the higher the association degree is, the more important the equipment operation parameter features are, and the equipment operation parameter features with the highest association degree are taken as importance features.
In one embodiment, calculating the weight value of the importance feature using the information gain includes the steps of:
s41, collecting a sample data set of equipment operation parameters;
s42, calculating information entropy according to target variables in the sample data set, and measuring uncertainty of the target variables;
specifically, the target variable refers to an operating state or an operating parameter of the device.
S43, for each feature, calculating the information gain between the feature and the target variable;
s44, taking the information gain value as a characteristic weight value, and carrying out normalization processing on the weight value;
specifically, normalization (Normalization): the normalization processing is to unify the dimensions of the data, so that the comparability among different data is realized, and the normalization processing is performed on the prediction result, so that the influence of the dimensions of the data is eliminated, and the subsequent weighting processing and summation calculation are facilitated.
S45, analyzing the normalized weight value, and knowing the importance degree of the operating parameter characteristics of the equipment.
In one embodiment, performing cluster analysis on devices based on weight values, marking a device region where an abnormal change exists as a potential problem region includes the steps of:
s51, taking the feature quantity of the importance features as the quantity of clustering centers, clustering the equipment areas by utilizing an improved K-Means clustering algorithm, and dividing the equipment areas into a plurality of groups;
specifically, the method for clustering the equipment areas by using the improved K-Means clustering algorithm and dividing the equipment areas into a plurality of groups comprises the following steps of:
s511, calculating noise measurement indexes of the data objects for the clustered device region data sets;
s512, regarding the data object, if the noise measurement index is larger than a preset threshold value, taking the data object as an isolated point of the data set;
s513, deleting the isolated points or leading out the isolated points to an abnormal value list to obtain a new equipment area data set X;
s514, randomly selecting K data objects from the new equipment area data set X as initial clustering centers C1, C2, … and C k ;
S515, calculating the distance between the data object and each cluster center according to the initial cluster center, and distributing the data object to the class where the nearest cluster center is located;
s516, distributing the data object to the nearest class, and calculating the average position of the data object as a new clustering center;
s517, if the new cluster center is the same as the cluster center in the step S516, ending the algorithm, otherwise, replacing the old cluster center with the new cluster center, and repeating the steps S515-S516;
s518, presetting an iteration number threshold, stopping iteration when the iteration number reaches the threshold, and outputting a final clustering result.
S52, extracting data points from a plurality of groups, calculating local outlier factors of the data points by utilizing a local outlier detection algorithm, and judging whether the data points are outliers;
specifically, the local outlier detection algorithm (Local Outlier Factor, abbreviated as LOF) is an algorithm for anomaly detection, and unlike many other outlier detection algorithms, the LOF considers the local characteristics of neighboring points, so it can more accurately identify outliers in various density areas, and in the LOF algorithm, the LOF score of an object is greater than 1, indicating that the object is more likely to be an outlier than its neighbors.
S53, counting the number of outliers in each group, and calculating the proportion of the outliers;
specifically, the proportion of outliers calculated is obtained by dividing the number of outliers by the total number of points in the cluster.
S54, if the proportion of the outliers in the group exceeds a preset threshold, judging that the group has abnormal change, and marking the equipment area with abnormal change as a potential problem area.
In one embodiment, the improved K-Means clustering algorithm is calculated as:
;
wherein,nrepresenting the number of data points in the cluster;
da feature quantity expressed as an importance feature;
V{S,Hdenoted as noise measure;
Sirepresent the firstiClustering;
X ih expressed in clustersS i Middle (f)hFeature vectors of data points;
X jh expressed in whole dataConcentrate all data points at the firsthAverage over individual features;
Hrepresented as a threshold;
jrepresenting a data point in the data.
Specifically, a noise metricV{S,HNoise or isolated point data is prone to interfere with the clustering effect, so noise metrics are added to improve the original K-Means clustering algorithm.
In one embodiment, constructing a device fault prediction model in the cloud using beamforming and time series analysis based on potential problem areas comprises the steps of:
s61, analyzing whether trend items with longer periods exist in the operation parameters of the equipment by adopting a polynomial regression model;
s62, if the trend item exists, removing the trend item with a longer period;
s63, if the periodic variation exists in the operation parameters of the equipment, continuously analyzing the periodic variation;
s64, performing a beam forming method on the equipment operation parameters after removing the trend item according to the analysis result to obtain the amplitude and the phase of each frequency component;
specifically, the amplitude represents the amount of change in the operating parameter of the device at a certain frequency, and the phase represents the time-lag characteristic of the change in the operating parameter of the device.
S65, judging whether each frequency component is obvious or not by using saliency test, and extracting a salient period term to construct a period term model;
s66, regarding residual errors after eliminating trend items and period items as random changes, and constructing a residual error prediction model;
and S67, superposing the polynomial regression model, the periodic term model and the residual prediction model to obtain the equipment fault prediction model.
Specifically, the significance test: the significance test is a statistical method for checking whether there is a significant difference between the observed data and a certain hypothesis. In time series analysis, a significance test is often used to determine if periodic components in the data are statistically significant. The results of the significance test are generally represented by p-values, with smaller p-values indicating more significant differences between the observed data and the hypothesis.
Specifically, the periodic term model: the periodic term model is mainly used to describe periodic components in time series data. In device failure prediction, a periodic term model may help capture periodic changes in device failure occurring on different time scales within a day, within a week, etc. Methods of constructing the periodic term model include fourier analysis, periodic regression, and the like.
Specifically, the trend term and the period term are eliminated: in time series analysis, data can generally be decomposed into trend terms, period terms, and random variations (residuals). By eliminating trend and period terms, the regularity component in the data can be culled, thereby focusing better on random variations. Methods of eliminating trend terms and period terms include differencing, filtering, and the like.
Specifically, the residual prediction model: the residual prediction model is used to describe random variations (residuals) in the time series data. After eliminating the trend term and the period term, the residual is regarded as a random variation, and the fluctuation at the next moment can be estimated by constructing a residual prediction model. Common residual prediction models include autoregressive moving average (ARIMA), exponential smoothing (ETS), and the like.
In one embodiment, predicting the occurrence of the equipment failure at the next moment by using the equipment failure prediction model and the weight value in the cloud comprises the following steps:
s71, predicting equipment operation parameters at the next moment through an equipment failure prediction model;
s72, carrying out corresponding weighting processing on the predicted equipment operation parameters and the weight values to obtain comprehensive prediction output of equipment failure occurrence.
According to another embodiment of the present invention, as shown in fig. 2, there is further provided a remote monitoring system based on an internet cloud, including:
the data acquisition and pretreatment module 1 is used for setting a plurality of sensors on equipment to be monitored, acquiring equipment operation parameters and carrying out pretreatment;
the feature extraction module 2 is used for extracting features from the preprocessing result and uploading the extracted features to the cloud for storage through the Internet;
the feature analysis module 3 is used for carrying out feature importance analysis on the extracted feature data and screening out importance features with the greatest influence on equipment abnormality early warning judgment;
a weight calculation module 4 for calculating a weight value of the importance feature using the information gain;
the abnormal change screening module 5 is used for carrying out cluster analysis on the equipment based on the weight value and marking the equipment area with abnormal change as a potential problem area;
the equipment fault prediction model construction module 6 is used for constructing an equipment fault prediction model in the cloud by utilizing a beam forming method and a time sequence analysis method based on the potential problem area;
the equipment fault prediction module 7 is used for predicting equipment fault occurrence at the next moment by using the equipment fault prediction model and the weight value in the cloud;
the acquired data are connected with the preprocessing module 1 through the feature extraction module 2 and the feature analysis module 3, the feature analysis module 3 is connected with the abnormal change screening module 5 through the weight calculation module 4, and the abnormal change screening module 5 is connected with the equipment fault prediction module 7 through the equipment fault prediction model construction module 6.
In summary, by means of the technical scheme, the device fault prediction model is built in the cloud by utilizing the beam forming method and the time sequence analysis method, so that the running trend of the device can be dynamically monitored and predicted, the tiny change of the performance of the device can be timely found, the device can be prevented and optimized before the fault occurs, the parameters of the prediction model can be dynamically adjusted according to the running state and environmental factors of the device, the intelligent decision is realized, the local resources can be greatly saved, the system performance and efficiency are improved, the running efficiency and reliability of the device are improved, and the running cost of the device is reduced. According to the invention, through an improved K-Means clustering algorithm, abnormal equipment can be effectively distinguished and marked as a potential problem area, so that equipment possibly having problems can be rapidly and effectively identified, and the equipment can be processed in time, thereby improving the efficiency of equipment management and maintenance and reducing equipment faults.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. The remote monitoring method based on the Internet cloud is characterized by comprising the following steps of:
s1, setting a plurality of sensors on equipment to be monitored, acquiring equipment operation parameters, and preprocessing;
s2, extracting features from the preprocessing result, and uploading the extracted features to a cloud for storage through the Internet;
s3, analyzing the feature importance of the extracted feature data, and screening out the importance feature with the greatest influence on the equipment abnormality early warning judgment;
s4, calculating a weight value of the importance feature by utilizing the information gain;
s5, carrying out cluster analysis on the equipment based on the weight value, and marking the equipment area with abnormal change as a potential problem area;
s6, constructing an equipment fault prediction model in the cloud by utilizing a beam forming method and a time sequence analysis method based on the potential problem area;
and S7, predicting the occurrence of the equipment fault at the next moment by using the equipment fault prediction model and the weight value in the cloud.
2. The remote monitoring method based on the internet cloud as claimed in claim 1, wherein the steps of setting a plurality of sensors on the equipment to be monitored, obtaining sensor data, and preprocessing include the following steps:
s11, collecting repeated data, missing values and abnormal values of equipment operation parameters acquired by each sensor, and denoising, filtering and smoothing the repeated data, the missing values and the abnormal values;
s12, connecting unprocessed data rows in the equipment operation parameters to generate a new data table, associating different data tables through external key values to generate a complete data table, and obtaining an accurate data set;
s13, determining external key relations among different data sets, connecting data in different data tables with each other according to requirements, creating a new data table, and associating through specified external key values;
s14, connecting the data tables to be joined together through the JOIN operators in the SQL sentences, and ensuring that the integrity constraint of the data is satisfied when the joining is carried out;
and S15, after the connection is completed, inserting test data to check whether the connection result is correct, so as to ensure that the connection result can be correctly identified and correlated, and obtaining accurate data of the operation parameters of the equipment.
3. The remote monitoring method based on the internet cloud according to claim 1, wherein the steps of extracting the features from the preprocessing result and uploading the extracted features to the cloud for storage through the internet comprise the following steps:
s21, fusing accurate data of equipment operation parameters into the same data set by using a principal component analysis method;
s22, extracting relevant characteristics from the fused data set to obtain equipment operation parameter characteristics;
s23, uploading the obtained equipment operation parameter characteristics to a cloud end through the Internet for storage.
4. The remote monitoring method based on the internet cloud as claimed in claim 1, wherein the feature importance analysis is performed on the extracted feature data, and the step of screening out the importance feature having the greatest influence on the equipment abnormality early warning judgment comprises the following steps:
s31, selecting and judging a characteristic evaluation index of the equipment abnormal behavior, and calculating the association degree between the equipment operation parameter characteristic and the equipment abnormal behavior based on the selected characteristic evaluation index;
s32, sorting the equipment operation parameter features based on the calculated association degree, wherein the higher the association degree is, the more important the equipment operation parameter features are, and the equipment operation parameter features with the highest association degree are taken as importance features.
5. The remote monitoring method based on the internet cloud as claimed in claim 1, wherein the calculating the weight value of the importance feature by using the information gain comprises the following steps:
s41, collecting a sample data set of equipment operation parameters;
s42, calculating information entropy according to target variables in the sample data set, and measuring uncertainty of the target variables;
s43, for each feature, calculating the information gain between the feature and the target variable;
s44, taking the information gain value as a characteristic weight value, and carrying out normalization processing on the weight value;
s45, analyzing the normalized weight value, and knowing the importance degree of the operating parameter characteristics of the equipment.
6. The remote monitoring method based on the internet cloud as set forth in claim 1, wherein the clustering analysis is performed on the equipment based on the weight value, and the marking of the equipment area with abnormal change as the potential problem area comprises the following steps:
s51, taking the feature quantity of the importance features as the quantity of clustering centers, clustering the equipment areas by utilizing an improved K-Means clustering algorithm, and dividing the equipment areas into a plurality of groups;
s52, extracting data points from a plurality of groups, calculating local outlier factors of the data points by utilizing a local outlier detection algorithm, and judging whether the data points are outliers;
s53, counting the number of outliers in each group, and calculating the proportion of the outliers;
s54, if the proportion of the outliers in the group exceeds a preset threshold, judging that the group has abnormal change, and marking the equipment area with abnormal change as a potential problem area.
7. The remote monitoring method based on the internet cloud as claimed in claim 6, wherein the calculation formula of the improved K-Means clustering algorithm is as follows:;
wherein,nrepresenting the number of data points in the cluster;
da feature quantity expressed as an importance feature;
V{S,Hdenoted as noise measure;
Sirepresent the firstiClustering;
X ih expressed in clustersS i Middle (f)hFeature vectors of data points;
X jh representing all data points in the whole data set at the firsthAverage over individual features;
Hrepresented as a threshold;
jrepresenting a data point in the data.
8. The remote monitoring method based on the internet cloud as claimed in claim 1, wherein the constructing the equipment fault prediction model in the cloud based on the potential problem area by using a beamforming method and a time series analysis method comprises the following steps:
s61, analyzing whether trend items with longer periods exist in the operation parameters of the equipment by adopting a polynomial regression model;
s62, if the trend item exists, removing the trend item with a longer period;
s63, if the periodic variation exists in the operation parameters of the equipment, continuously analyzing the periodic variation;
s64, performing a beam forming method on the equipment operation parameters after removing the trend item according to the analysis result to obtain the amplitude and the phase of each frequency component;
s65, judging whether each frequency component is obvious or not by using saliency test, and extracting a salient period term to construct a period term model;
s66, regarding residual errors after eliminating trend items and period items as random changes, and constructing a residual error prediction model;
and S67, superposing the polynomial regression model, the periodic term model and the residual prediction model to obtain the equipment fault prediction model.
9. The remote monitoring method based on the internet cloud as claimed in claim 1, wherein the predicting the occurrence of the equipment failure at the next moment by using the equipment failure prediction model and the weight value in the cloud comprises the following steps:
s71, predicting equipment operation parameters at the next moment through an equipment failure prediction model;
s72, carrying out corresponding weighting processing on the predicted equipment operation parameters and the weight values to obtain comprehensive prediction output of equipment failure occurrence.
10. An internet cloud based remote monitoring system for implementing the internet cloud based remote monitoring method of any one of claims 1 to 9, wherein the internet cloud based remote monitoring system comprises:
the data acquisition and pretreatment module is used for setting a plurality of sensors on equipment to be monitored, acquiring equipment operation parameters and carrying out pretreatment;
the feature extraction module is used for extracting features from the preprocessing result and uploading the extracted features to the cloud for storage through the Internet;
the feature analysis module is used for carrying out feature importance analysis on the extracted feature data and screening out importance features with the greatest influence on equipment abnormality early warning judgment;
the weight calculation module is used for calculating the weight value of the importance characteristic by utilizing the information gain;
the abnormal change screening module is used for carrying out cluster analysis on the equipment based on the weight value and marking the equipment area with abnormal change as a potential problem area;
the equipment fault prediction model construction module is used for constructing an equipment fault prediction model in the cloud by utilizing a beam forming method and a time sequence analysis method based on the potential problem area;
the equipment fault prediction module is used for predicting equipment fault occurrence at the next moment by using the equipment fault prediction model and the weight value in the cloud;
the device comprises a preprocessing module, a weight calculation module, an abnormal change screening module, an equipment fault prediction model construction module, a device fault prediction model, a characteristic analysis module, a weight calculation module, an equipment fault prediction model, a weight calculation module and an abnormal change screening module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311328782.7A CN117076258A (en) | 2023-10-15 | 2023-10-15 | Remote monitoring method and system based on Internet cloud |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311328782.7A CN117076258A (en) | 2023-10-15 | 2023-10-15 | Remote monitoring method and system based on Internet cloud |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117076258A true CN117076258A (en) | 2023-11-17 |
Family
ID=88717387
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311328782.7A Pending CN117076258A (en) | 2023-10-15 | 2023-10-15 | Remote monitoring method and system based on Internet cloud |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117076258A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117308275A (en) * | 2023-11-28 | 2023-12-29 | 江苏嘉尚环保科技有限公司 | Temperature difference-based pipeline connection abnormality detection method and system |
CN117851787A (en) * | 2024-03-01 | 2024-04-09 | 江苏森讯达智能科技有限公司 | Intelligent interaction control method and system for bathroom products |
CN117932520A (en) * | 2024-03-20 | 2024-04-26 | 史瑞美(厦门)科技有限公司 | Solid biological waste treatment equipment monitoring method based on data identification |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106549813A (en) * | 2015-09-16 | 2017-03-29 | 中兴通讯股份有限公司 | A kind of appraisal procedure and system of network performance |
CN113127716A (en) * | 2021-04-29 | 2021-07-16 | 南京大学 | Sentiment time sequence anomaly detection method based on saliency map |
CN116881834A (en) * | 2023-09-08 | 2023-10-13 | 泰州市银杏舞台机械工程有限公司 | Stage load monitoring and early warning method based on stage deformation analysis |
-
2023
- 2023-10-15 CN CN202311328782.7A patent/CN117076258A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106549813A (en) * | 2015-09-16 | 2017-03-29 | 中兴通讯股份有限公司 | A kind of appraisal procedure and system of network performance |
CN113127716A (en) * | 2021-04-29 | 2021-07-16 | 南京大学 | Sentiment time sequence anomaly detection method based on saliency map |
CN116881834A (en) * | 2023-09-08 | 2023-10-13 | 泰州市银杏舞台机械工程有限公司 | Stage load monitoring and early warning method based on stage deformation analysis |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117308275A (en) * | 2023-11-28 | 2023-12-29 | 江苏嘉尚环保科技有限公司 | Temperature difference-based pipeline connection abnormality detection method and system |
CN117308275B (en) * | 2023-11-28 | 2024-02-06 | 江苏嘉尚环保科技有限公司 | Temperature difference-based pipeline connection abnormality detection method and system |
CN117851787A (en) * | 2024-03-01 | 2024-04-09 | 江苏森讯达智能科技有限公司 | Intelligent interaction control method and system for bathroom products |
CN117932520A (en) * | 2024-03-20 | 2024-04-26 | 史瑞美(厦门)科技有限公司 | Solid biological waste treatment equipment monitoring method based on data identification |
CN117932520B (en) * | 2024-03-20 | 2024-06-07 | 史瑞美(厦门)科技有限公司 | Solid biological waste treatment equipment monitoring method based on data identification |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117076258A (en) | Remote monitoring method and system based on Internet cloud | |
US10901832B2 (en) | System for maintenance recommendation based on failure prediction | |
CN105677538B (en) | A kind of cloud computing system self-adaptive monitoring method based on failure predication | |
CN101771758A (en) | Dynamic determine method for normal fluctuation range of performance index value and device thereof | |
US20100030521A1 (en) | Method for analyzing and classifying process data | |
TWI662424B (en) | Selection method of leading auxiliary parameters and method for pre-diagnosis of equipment maintenance by combining key parameters and leading auxiliary parameters | |
CN110688617B (en) | Fan vibration abnormity detection method and device | |
CN114267178B (en) | Intelligent operation maintenance method and device for station | |
CN117308275B (en) | Temperature difference-based pipeline connection abnormality detection method and system | |
CN113568774A (en) | Real-time anomaly detection method for multi-dimensional time sequence data by using unsupervised deep neural network | |
CN111756560A (en) | Data processing method, device and storage medium | |
CN115800272A (en) | Power grid fault analysis method, system, terminal and medium based on topology identification | |
KR101960755B1 (en) | Method and apparatus of generating unacquired power data | |
US20230034061A1 (en) | Method for managing proper operation of base station and system applying the method | |
JP6602256B2 (en) | Program, apparatus and method capable of detecting time series change point based on cross-correlation | |
CN117093461A (en) | Method, system, equipment and storage medium for time delay detection and analysis | |
CN108362957B (en) | Equipment fault diagnosis method and device, storage medium and electronic equipment | |
KR101884908B1 (en) | Big Data Analytics Based Reliability Prediction Apparatus | |
JP6587950B2 (en) | Program, apparatus, and method capable of detecting time series change point by scalar feature | |
CN111814331B (en) | Multi-point equipment residual service life prediction method under complex condition | |
CN117278591A (en) | Park abnormity alarm system based on cloud platform | |
CN117435908A (en) | Multi-fault feature extraction method for rotary machine | |
CN114938339B (en) | Data processing method and related device | |
RU2703874C1 (en) | Method of monitoring and predicting operation of a gas turbine plant using a matrix of defects | |
CN115904916A (en) | Hard disk failure prediction method and device, electronic equipment and storage medium |
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 |