CN115834433A - Data processing method and system based on Internet of things technology - Google Patents
Data processing method and system based on Internet of things technology Download PDFInfo
- Publication number
- CN115834433A CN115834433A CN202310128375.5A CN202310128375A CN115834433A CN 115834433 A CN115834433 A CN 115834433A CN 202310128375 A CN202310128375 A CN 202310128375A CN 115834433 A CN115834433 A CN 115834433A
- Authority
- CN
- China
- Prior art keywords
- internet
- working power
- matrix
- feature
- things
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The data processing method and system based on the Internet of things technology are disclosed, and the Internet of things technology is utilized to directly collect relevant data of monitored Internet of things terminal equipment to carry out performance monitoring and fault early warning on the monitored Internet of things terminal equipment. Specifically, performing one-dimensional convolution coding and full-connection coding on the working power values of the internet of things terminal devices at a plurality of preset time points in a preset time period to extract high-dimensional implicit characteristics of the working power of the internet of things terminal devices at the time points and high-dimensional implicit association mode characteristics associated with the working power of the internet of things terminal devices at the time points; and high-dimensional local implicit associated features of the terminal equipment of the internet of things on the spatial topological dimension are mined to optimize the feature expression of the working features of the terminal equipment of the internet of things, and the classification accuracy is improved. Therefore, the state of the terminal equipment of the Internet of things is detected based on the technology of the Internet of things.
Description
Technical Field
The present application relates to the field of data processing, and more particularly, to a data processing method and system based on internet of things.
Background
The internet of things is a network which is based on information carriers such as the internet, a traditional telecommunication network and the like and enables all common physical objects which can be independently addressed to realize interconnection. Common object equipment, autonomous terminal interconnection and pervasive service intellectualization are three important characteristics. With the continuous perfection of the management platform of the internet of things, the types of terminals are continuously enriched and the number of the terminals is also increased sharply, and the following problems are that: in the face of huge internet of things terminal equipment, a large amount of manpower and material resources are consumed to carry out fault detection and maintenance on the internet of things terminal equipment one by one.
Therefore, a data processing scheme based on the internet of things technology is expected, and the intelligent monitoring and fault early warning of the device performance of the terminal device of the internet of things can be realized.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a data processing method and system based on the technology of the Internet of things, which directly collect relevant data of monitored terminal equipment of the Internet of things to carry out performance monitoring and fault early warning on the monitored terminal equipment of the Internet of things by utilizing the technology of the Internet of things. Specifically, performing one-dimensional convolution coding and full-connection coding on the working power values of the internet of things terminal devices at a plurality of preset time points in a preset time period to extract high-dimensional implicit characteristics of the working power of the internet of things terminal devices at the time points and high-dimensional implicit association mode characteristics associated with the working power of the internet of things terminal devices at the time points; and high-dimensional local implicit associated features of the terminal equipment of the internet of things on the spatial topological dimension are mined to optimize the feature expression of the working features of the terminal equipment of the internet of things, and the classification accuracy is improved. Therefore, the state of the terminal equipment of the Internet of things is detected based on the technology of the Internet of things.
According to one aspect of the application, a data processing method based on the technology of the internet of things is provided, and comprises the following steps:
the method comprises the steps of obtaining working power values of a plurality of internet of things terminal devices at a plurality of preset time points in a preset time period;
respectively arranging the working power values of a plurality of preset time points of each Internet of things terminal device in a preset time period into working power input vectors according to the time dimension, and then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer;
constructing a spatial topological matrix of the plurality of internet of things terminal devices, wherein the value of each position on the off-diagonal position of the spatial topological matrix is the distance between two corresponding internet of things terminal devices;
enabling the spatial topological matrix to pass through a convolutional neural network model serving as a feature extractor to obtain a spatial topological feature matrix;
the working power eigenvectors are arranged in a two-dimensional mode to obtain a working power global eigenvector matrix;
passing the working power global feature matrix and the spatial topological feature matrix through a graph neural network model to obtain a topological working power global feature matrix;
performing characteristic polymerization degree optimization among row vectors on the topology working power global characteristic matrix to obtain an optimized topology working power global characteristic matrix;
extracting row vectors of the terminal equipment of the Internet of things to be detected from the optimized topology working power global feature matrix to serve as classification feature vectors; and
and passing the classified characteristic vectors through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the state of the to-be-detected Internet of things terminal equipment is abnormal or not.
In the data processing method based on the internet of things technology, after the working power values of the internet of things terminal devices at a plurality of predetermined time points in a predetermined time period are respectively arranged as working power input vectors according to a time dimension, a plurality of working power characteristic vectors are obtained through a time sequence encoder including a one-dimensional convolution layer and a full connection layer, including: performing one-dimensional convolution encoding on the working power input vector by using a one-dimensional convolution layer of the time sequence encoder according to the following formula so as to extract high-dimensional implicit correlation characteristics among characteristic values of all positions in the working power input vector, wherein the formula is as follows:
wherein a is the width of the convolution kernel in the x direction,
Is a convolution kernel parameter vector,
Is a matrix of local vectors operating with a convolution kernel, w is the size of the convolution kernel,
representing the operating power input vector and,
representing a one-dimensional convolution coding operation on the working power input vector; and using a full connection layer of the time sequence encoder to perform full connection encoding on the high-dimensional implicit associated features by using the following formula so as to extract the high-dimensional implicit features of feature values of all positions in the high-dimensional implicit associated features, wherein the formula is as follows:
wherein
Is the high-dimensional implicit associative feature that is,
is an outputThe vector of the vector is then calculated,
is a matrix of the weights that is,
is a vector of the offset to the offset,
representing a matrix multiplication.
In the data processing method based on the internet of things technology, the obtaining the spatial topological characteristic matrix by using the spatial topological matrix through a convolutional neural network model as a characteristic extractor includes: using the layers of the convolutional neural network model in forward pass of the layers respectively: performing convolution processing on input data to obtain a convolution characteristic diagram; performing mean pooling based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; and the output of the last layer of the convolutional neural network model is the spatial topological characteristic matrix, and the input of the first layer of the convolutional neural network model is the spatial topological matrix.
In the data processing method based on the internet of things technology, the convolutional neural network model serving as the feature extractor is a depth residual error network.
In the data processing method based on the internet of things technology, the deep residual error network is ResNet 150.
In the data processing method based on the internet of things technology, the optimizing the inter-row vector feature polymerization degree of the topology working power global feature matrix to obtain an optimized topology working power global feature matrix includes: performing characteristic polymerization degree optimization among row vectors on the topology working power global characteristic matrix according to the following formula to obtain an optimized topology working power global characteristic matrix; wherein the formula is:
wherein the content of the first and second substances,
is each topology operating power global eigenvector of said topology operating power global eigenvector matrix,
is the topology working power global feature vector of the topology working power global feature matrix and the topology working power global feature vector
A distance therebetween, i.e.
Less than a predetermined threshold value (
) The topology operating power global feature vector of (a),
in order to weight the hyper-parameters,
it is shown that the difference is made by position,
an exponential operation of a vector representing a calculation of a natural exponential function value raised to a power of a feature value of each position in the vector is represented.
In the data processing method based on the internet of things technology, the step of passing the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the state of the terminal device of the internet of things to be detected is abnormal or not includes: performing full-concatenation coding on the classification feature vectors by using a full-concatenation layer of the classifier to obtain coded classification feature vectors; and inputting the encoding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
According to another aspect of the present application, there is provided a data processing system based on internet of things technology, including:
the device power acquisition module is used for acquiring the working power values of a plurality of Internet of things terminal devices at a plurality of preset time points in a preset time period;
the power characteristic extraction module is used for arranging working power values of a plurality of preset time points of each Internet of things terminal device in a preset time period into working power input vectors according to a time dimension, and then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer;
the space topology construction module is used for constructing a space topology matrix of the plurality of internet of things terminal devices, and the value of each position on the off-diagonal position of the space topology matrix is the distance between two corresponding internet of things terminal devices;
the topological characteristic extraction module is used for enabling the space topological matrix to pass through a convolutional neural network model serving as a characteristic extractor so as to obtain a space topological characteristic matrix;
the global arrangement module is used for carrying out two-dimensional arrangement on the working power eigenvectors to obtain a working power global eigenvector matrix;
the graph neural network coding module is used for enabling the working power global characteristic matrix and the spatial topological characteristic matrix to pass through a graph neural network model so as to obtain a topological working power global characteristic matrix;
the inter-vector characteristic polymerization degree optimization module is used for performing inter-row vector characteristic polymerization degree optimization on the topology working power global characteristic matrix to obtain an optimized topology working power global characteristic matrix;
the extraction module of the to-be-detected Internet of things terminal equipment is used for extracting row vectors of the to-be-detected Internet of things terminal equipment from the optimized topology working power global feature matrix as classification feature vectors; and
and the detection result generation module is used for enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the state of the to-be-detected Internet of things terminal equipment is abnormal or not.
In the data processing system based on the internet of things, the power feature extraction module is further configured to: performing one-dimensional convolution encoding on the working power input vector by using a one-dimensional convolution layer of the time sequence encoder according to the following formula so as to extract high-dimensional implicit correlation characteristics among characteristic values of all positions in the working power input vector, wherein the formula is as follows:
wherein a is the width of the convolution kernel in the x direction,
Is a convolution kernel parameter vector,
Is a matrix of local vectors operating with a convolution kernel, w is the size of the convolution kernel,
representing the operating power input vector and,
representing a one-dimensional convolution coding operation on the working power input vector; and using a full connection layer of the time sequence encoder to perform full connection encoding on the high-dimensional implicit associated features by using the following formula so as to extract the high-dimensional implicit features of feature values of all positions in the high-dimensional implicit associated features, wherein the formula is as follows:
wherein
Is the high-dimensional implicit associative feature that is,
is the output vector of the output vector,
is a weightThe matrix is a matrix of a plurality of matrices,
is a vector of the offset to the offset,
representing a matrix multiplication.
In the data processing system based on the internet of things, the topological feature extraction module is further configured to: using the layers of the convolutional neural network model in forward pass of the layers respectively: performing convolution processing on input data to obtain a convolution characteristic diagram; performing mean pooling based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; and the output of the last layer of the convolutional neural network model is the spatial topological characteristic matrix, and the input of the first layer of the convolutional neural network model is the spatial topological matrix.
In the data processing system based on the internet of things technology, the convolutional neural network model serving as the feature extractor is a depth residual error network.
In the data processing system based on the internet of things technology, the deep residual network is ResNet 150.
In the data processing system based on the internet of things, the inter-vector feature polymerization degree optimization module is further configured to: performing characteristic polymerization degree optimization among row vectors on the topology working power global characteristic matrix according to the following formula to obtain an optimized topology working power global characteristic matrix; wherein the formula is:
wherein the content of the first and second substances,
is each topology operating power global eigenvector of said topology operating power global eigenvector matrix,
is the topology working power global feature vector of the topology working power global feature matrix and the topology working power global feature vector
Is a distance therebetween, i.e.
Less than a predetermined threshold value (
) The topology operating power global feature vector of (a),
in order to weight the hyper-parameters,
it is shown that the difference is made by position,
an exponential operation of a vector representing a calculation of a natural exponential function value raised to a power of a feature value of each position in the vector is represented.
In the data processing system based on the internet of things, the detection result generation module is further configured to: performing full-joint coding on the classification feature vectors by using a full-joint layer of the classifier to obtain coded classification feature vectors; and inputting the encoding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory in which computer program instructions are stored, which, when executed by the processor, cause the processor to perform the internet of things technology based data processing method as described above.
According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions, which, when executed by a processor, cause the processor to perform the method of data processing based on internet of things as described above.
Compared with the prior art, the data processing method and system based on the internet of things technology provided by the application directly collect the relevant data of the monitored internet of things terminal equipment to carry out performance monitoring and fault early warning on the monitored internet of things terminal equipment by using the internet of things technology. Specifically, performing one-dimensional convolution coding and full-connection coding on the working power values of the internet of things terminal devices at a plurality of preset time points in a preset time period to extract high-dimensional implicit characteristics of the working power of the internet of things terminal devices at the time points and high-dimensional implicit association mode characteristics associated with the working power of the internet of things terminal devices at the time points; and high-dimensional local implicit associated features of the terminal equipment of the internet of things on the spatial topological dimension are mined to optimize the feature expression of the working features of the terminal equipment of the internet of things, and the classification accuracy is improved. Therefore, the state of the terminal equipment of the Internet of things is detected based on the technology of the Internet of things.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally represent like parts or steps.
Fig. 1 is a flowchart of a data processing method based on internet of things according to an embodiment of the present application.
Fig. 2 is an architecture diagram of a data processing method based on internet of things according to an embodiment of the present application.
Fig. 3 is a flowchart of obtaining a classification result by passing the classification feature vector through a classifier in the data processing method based on the internet of things according to the embodiment of the present application.
Fig. 4 is a block diagram of a data processing system based on internet of things technology according to an embodiment of the application.
Fig. 5 is a block diagram of an electronic device according to an embodiment of the application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and that the present application is not limited by the example embodiments described herein.
Summary of the application
As described above, with the continuous improvement of the internet of things management platform, the types of terminals are continuously abundant and the number of terminals is also sharply increased, and the following problems are that: in the face of huge internet of things terminal equipment, a large amount of manpower and material resources are consumed to carry out fault detection and maintenance on the internet of things terminal equipment one by one. For example, in an intelligent cell phone production workshop, in order to improve the degree of intellectualization and automation, various internet of things terminal devices are configured, which leads to a sudden increase in difficulty in fault detection and maintenance of the internet of things terminal devices. Therefore, a data processing scheme based on the internet of things technology is expected, and the intelligent monitoring and fault early warning of the device performance of the terminal device of the internet of things can be realized.
Specifically, in the technical scheme of this application, because of the thing networking terminal equipment monitored is in the thing networking, consequently can directly gather monitored thing networking terminal equipment's relevant data comes to carry out performance monitoring and trouble early warning to it. However, in the physical network device system, the terminal devices of the internet of things are not independent from each other, and if performance monitoring and fault early warning are performed only by relying on the individual data of the terminal devices of the internet of things, relatively large errors and deviations are generated.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks also exhibit a level close to or even exceeding that of humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like. The deep learning and the development of the neural network provide a new solution for the performance monitoring and the fault early warning of the terminal equipment of the Internet of things.
Specifically, firstly, working power values of a plurality of internet of things terminal devices at a plurality of predetermined time points in a predetermined time period are obtained. That is, in the technical scheme of this application, carry out real-time supervision to each thing networking terminal equipment's operating power through thing networking technology, and based on the chronogenesis characteristic of thing networking terminal equipment's operating power carries out performance monitoring and trouble early warning to it.
And then, respectively arranging the working power values of the terminal equipment of the Internet of things at a plurality of preset time points in a preset time period into working power input vectors according to a time dimension, and then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer. Namely, the time sequence encoder comprising the one-dimensional convolutional layer and the full connection layer is used for carrying out one-dimensional convolutional coding and full connection coding on the time sequence vector of the working power of each internet of things terminal device so as to extract high-dimensional implicit characteristics of the working power of each internet of things terminal device at each time point and high-dimensional implicit association mode characteristics associated with the working power of each time point. In one specific example, the time-sequential encoder is composed of one-dimensional convolutional layers and fully-connected layers which are alternately arranged.
Particularly, in the technical scheme of the application, it is considered that the terminal devices of the internet of things are not independent from each other, and in the layout of the terminal devices of the internet of things, the coordination probability between the terminal devices of the internet of things at adjacent distances is higher, that is, the association mode between the working power characteristics between the terminal devices of the internet of things at relatively closer distances is deeper and more obvious. Based on this, in the technical scheme of this application, construct the spatial topology matrix of a plurality of thing networking terminal equipment, the value of each position on the off-diagonal position of spatial topology matrix is the distance between two corresponding thing networking terminal equipment, and will the spatial topology matrix is through the convolutional neural network model as the feature extractor in order to obtain the spatial topology feature matrix. Namely, the spatial distance topology of the plurality of internet of things terminal devices is constructed, and the high-dimensional local implicit associated features of the spatial distance topology of the plurality of internet of things terminal devices are extracted by using a convolutional neural network model with excellent performance in the local feature extraction field.
And then, the working power characteristic vectors of the terminal devices of the internet of things are used as nodes, the distances among the terminal devices of the internet of things are used as edges, the edges are used as graph structure data, and graph data are coded by utilizing a graph neural network model. Specifically, the working power feature vectors are arranged in a two-dimensional mode to obtain a working power global feature matrix, and the working power global feature matrix and the spatial topological feature matrix are used for obtaining a topological working power global feature matrix through a graph neural network model. Specifically, the graph neural network model processes the working power global feature matrix and the spatial topological feature matrix through learnable neural network parameters to obtain a feature expression containing irregular spatial topological features and working features of the internet of things terminal equipment. In the topology working power global feature matrix, each row vector corresponds to a topology working power global feature vector of each internet of things terminal device.
And then, extracting a row vector of the terminal equipment of the Internet of things to be detected from the topology working power global feature matrix as a classification feature vector, and enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the state of the terminal equipment of the Internet of things to be detected is abnormal or not. The method comprises the steps of extracting a topological working characteristic expression of the terminal equipment of the Internet of things to be detected, and determining a class probability label to which the topological working characteristic expression belongs by using a classifier.
Particularly, in the technical scheme of the application, when the working power global feature matrix and the spatial topology feature matrix are used for obtaining the topology working power global feature matrix through a graph neural network model, each topology working power global feature vector (for example, a row vector) of the topology working power global feature matrix expresses high-dimensional associated expression of working power associated features of a single internet of things terminal device under a spatial topology, and therefore, the problem of low feature polymerization degree may exist among the topology working power global feature vectors of the topology working power global feature matrix, which may affect the feature expression effect of the extracted classification feature vector, thereby affecting the accuracy of the classification result of the classification feature vector through a classifier.
Therefore, it is considered that if each topology working power global feature vector in the topology working power global feature vectors of the topology working power global feature matrix is regarded as a single node, the class probability feature polymerization degree between the nodes can be determined based on the topology structure of the whole node through the distance representation between the nodes, and thus, the inter-node class probability matching feature vector corresponding to each topology working power global feature vector is calculated and represented as:
That is, if will
As a node of the topology, then
Can be considered as being within the topology with said node
Connected nodes, i.e.
Representing nodes under a topology
And node
With edges in between. Therefore, the inter-node class probability matching feature vector is calculated to represent the interaction degree between the nodes and the adjacent nodes in the topological structure under the class probability, the topological working power global feature vector replacing the topological working power global feature matrix is rearranged into the topological working power global feature matrix, the class probability feature polymerization degree between all the nodes in the topological structure formed by multiple nodes can be improved, and the method is equivalent to applying an attention mechanism to the node features on the feature polymerization dimension based on the internal feature interaction, so that the feature polymerization degree between all the topological working power global feature vectors of the topological working power global feature matrix is improved, and the accuracy of the extracted classification result of the classification feature vector passing through the classifier is also improved.
Based on this, the application provides a data processing method based on the internet of things technology, which includes: the method comprises the steps of obtaining working power values of a plurality of internet of things terminal devices at a plurality of preset time points in a preset time period; respectively arranging the working power values of the internet of things terminal equipment at a plurality of preset time points in a preset time period into working power input vectors according to a time dimension, and then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer; constructing a spatial topological matrix of the plurality of internet of things terminal devices, wherein the value of each position on the off-diagonal position of the spatial topological matrix is the distance between two corresponding internet of things terminal devices; passing the spatial topological matrix through a convolutional neural network model serving as a feature extractor to obtain a spatial topological feature matrix; the working power eigenvectors are arranged in a two-dimensional mode to obtain a working power global eigenvector matrix; passing the working power global feature matrix and the spatial topological feature matrix through a graph neural network model to obtain a topological working power global feature matrix; performing inter-row vector feature polymerization degree optimization on the topology working power global feature matrix to obtain an optimized topology working power global feature matrix; extracting row vectors of the terminal equipment of the Internet of things to be detected from the optimized topology working power global feature matrix to serve as classification feature vectors; and enabling the classified characteristic vectors to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the state of the to-be-detected Internet of things terminal equipment is abnormal or not.
Having described the general principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.
Exemplary method
Fig. 1 is a flowchart of a data processing method based on internet of things according to an embodiment of the present application. As shown in fig. 1, a data processing method based on internet of things according to an embodiment of the present application includes: s110, obtaining working power values of a plurality of Internet of things terminal devices at a plurality of preset time points in a preset time period; s120, respectively arranging the working power values of the terminal equipment of the Internet of things at a plurality of preset time points in a preset time period into working power input vectors according to a time dimension, and then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer; s130, constructing a spatial topological matrix of the plurality of Internet of things terminal devices, wherein the value of each position on the off-diagonal position of the spatial topological matrix is the distance between two corresponding Internet of things terminal devices; s140, passing the spatial topological matrix through a convolutional neural network model serving as a feature extractor to obtain a spatial topological feature matrix; s150, performing two-dimensional arrangement on the plurality of working power characteristic vectors to obtain a working power global characteristic matrix; s160, passing the working power global feature matrix and the space topological feature matrix through a graph neural network model to obtain a topological working power global feature matrix; s170, optimizing the inter-row vector characteristic polymerization degree of the topology working power global characteristic matrix to obtain an optimized topology working power global characteristic matrix; s180, extracting a row vector of the terminal equipment of the Internet of things to be detected from the global characteristic matrix of the optimized topology working power as a classification characteristic vector; and S190, enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the state of the to-be-detected Internet of things terminal equipment is abnormal or not.
Fig. 2 is an architecture diagram of a data processing method based on internet of things according to an embodiment of the present application. As shown in fig. 2, in the framework, first, working power values of a plurality of internet of things terminal devices at a plurality of predetermined time points within a predetermined time period are obtained; then, respectively arranging the working power values of the internet of things terminal equipment at a plurality of preset time points in a preset time period into working power input vectors according to a time dimension, then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer, and meanwhile, constructing a spatial topological matrix of the plurality of internet of things terminal equipment, wherein the value of each position on the non-diagonal position of the spatial topological matrix is the distance between two corresponding internet of things terminal equipment; then, the spatial topological matrix is used as a convolutional neural network model of a feature extractor to obtain a spatial topological feature matrix; then, the working power eigenvectors are arranged in a two-dimensional mode to obtain a working power global eigenvector matrix; then, passing the working power global characteristic matrix and the space topological characteristic matrix through a graph neural network model to obtain a topological working power global characteristic matrix; then, optimizing the inter-row vector characteristic polymerization degree of the topology working power global characteristic matrix to obtain an optimized topology working power global characteristic matrix; extracting row vectors of the terminal equipment of the Internet of things to be detected from the optimized topology working power global feature matrix as classified feature vectors; and then, the classification feature vectors are processed by a classifier to obtain a classification result, and the classification result is used for indicating whether the state of the to-be-detected Internet of things terminal equipment is abnormal or not.
In step S110, working power values of a plurality of internet of things terminal devices at a plurality of predetermined time points in a predetermined time period are obtained. As described above, with the continuous improvement of the internet of things management platform, the types of terminals are continuously abundant and the number of terminals is also sharply increased, and the following problems are that: in the face of huge internet of things terminal equipment, a large amount of manpower and material resources are consumed to carry out fault detection and maintenance on the internet of things terminal equipment one by one. For example, in an intelligent cell phone production workshop, in order to improve the degree of intellectualization and automation, various internet of things terminal devices are configured, which leads to a sudden increase in difficulty in fault detection and maintenance of the internet of things terminal devices. Therefore, a data processing scheme based on the internet of things technology is expected, and the intelligent monitoring and fault early warning of the device performance of the terminal device of the internet of things can be realized.
Specifically, in the technical scheme of this application, because of the thing networking terminal equipment monitored is in the thing networking, consequently can directly gather monitored thing networking terminal equipment's relevant data comes to carry out performance monitoring and trouble early warning to it. However, in the physical network device system, the terminal devices of the internet of things are not independent from each other, and if performance monitoring and fault early warning are performed only by relying on the individual data of the terminal devices of the internet of things, relatively large errors and deviations are generated.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks also exhibit a level close to or even exceeding that of humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like. The deep learning and the development of the neural network provide a new solution for performance monitoring and fault early warning of the terminal equipment of the Internet of things.
Specifically, firstly, working power values of a plurality of internet of things terminal devices at a plurality of predetermined time points in a predetermined time period are obtained. That is, in the technical scheme of this application, carry out real-time supervision to each thing networking terminal equipment's operating power through internet of things, and based on thing networking terminal equipment's operating power's chronogenesis characteristic comes to carry out performance monitoring and trouble early warning to it.
In step S120, after the operating power values of the internet of things terminal devices at a plurality of predetermined time points in a predetermined time period are respectively arranged as operating power input vectors according to a time dimension, a plurality of operating power feature vectors are obtained through a time sequence encoder including a one-dimensional convolutional layer and a full link layer. Namely, the time sequence encoder comprising the one-dimensional convolutional layer and the full-link layer is used for performing one-dimensional convolutional coding and full-link coding on the time sequence vector of the working power of each internet of things terminal device so as to extract the high-dimensional implicit characteristics of the working power of each time point of each internet of things terminal device and the high-dimensional implicit association mode characteristics of association between the working powers of each time point. In one specific example, the time-sequential encoder is composed of one-dimensional convolutional layers and fully-connected layers which are alternately arranged.
Specifically, in this embodiment of the present application, first, a one-dimensional convolution layer of the time-series encoder is used to perform one-dimensional convolution encoding on the working power input vector by using the following formula to extract high-dimensional implicit correlation features between feature values of each position in the working power input vector, where the formula is:
wherein a is the width of the convolution kernel in the x direction,
Is a convolution kernel parameter vector,
Is a matrix of local vectors operating with a convolution kernel, w is the size of the convolution kernel,
representing the operating power input vector and,
representing a one-dimensional convolution coding operation on the working power input vector; then, using a full-connection layer of the time sequence encoder to perform full-connection encoding on the high-dimensional implicit associated features by using the following formula to extract the high-dimensional implicit features of feature values of each position in the high-dimensional implicit associated features, wherein the formula is as follows:
wherein
Is the high-dimensional implicit associative feature that is,
is the output vector of the output vector,
is a matrix of the weights that is,
is a vector of the offset to the offset,
representing a matrix multiplication.
It should be understood that the input of the sequential encoder is the working power input vector, the convolutional encoding operation is performed through the alternately arranged one-dimensional convolutional layers and fully-connected layers, and finally, the last layer of the sequential encoder outputs the working power feature vector.
In step S130, a spatial topology matrix of the plurality of internet of things terminal devices is constructed, where a value of each position on an off-diagonal position of the spatial topology matrix is a distance between two corresponding internet of things terminal devices. Particularly, in the technical scheme of the application, it is considered that the terminal devices of the internet of things are not independent from each other, and in the layout of the terminal devices of the internet of things, the coordination probability between the terminal devices of the internet of things at adjacent distances is higher, that is, the association mode between the working power characteristics between the terminal devices of the internet of things at relatively closer distances is deeper and more obvious.
Based on this, in the technical scheme of this application, construct the spatial topology matrix of a plurality of thing networking terminal device at first, the value of each position on the off-diagonal position of spatial topology matrix is the distance between two corresponding thing networking terminal device.
In step S140, the spatial topological matrix is passed through a convolutional neural network model as a feature extractor to obtain a spatial topological feature matrix. That is, after the spatial topological matrices of the plurality of internet of things terminal devices are obtained, the spatial topological matrices are used as a convolutional neural network model of a feature extractor to obtain spatial topological feature matrices. According to the technical scheme, space distance topologies of the plurality of Internet of things terminal devices are constructed, and high-dimensional local implicit associated features of the space distance topologies of the plurality of Internet of things terminal devices are extracted by using a convolutional neural network model with excellent performance in the field of local feature extraction.
Specifically, in this embodiment of the present application, the passing the spatial topology matrix through a convolutional neural network model as a feature extractor to obtain a spatial topology feature matrix includes: using the layers of the convolutional neural network model in forward pass of the layers respectively: performing convolution processing on input data to obtain a convolution characteristic diagram; performing mean pooling based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; and the output of the last layer of the convolutional neural network model is the spatial topological characteristic matrix, and the input of the first layer of the convolutional neural network model is the spatial topological matrix. In one particular embodiment of the present application, the convolutional neural network model as a feature extractor is a deep residual network, and more specifically, the deep residual network is ResNet 150.
In steps S150 and S160, the working power feature vectors are two-dimensionally arranged to obtain a working power global feature matrix, and the working power global feature matrix and the spatial topological feature matrix are passed through a neural network model to obtain a topological working power global feature matrix. Namely, the working power characteristic vectors of the terminal devices of the internet of things are used as nodes, the distances among the terminal devices of the internet of things are used as edges, the edges are used as graph structure data, and a graph neural network model is used for coding the graph data.
Specifically, firstly, the working power feature vectors are arranged in two dimensions to obtain a working power global feature matrix, and then the working power global feature matrix and the spatial topological feature matrix are processed through a graph neural network model to obtain a topological working power global feature matrix.
And the graph neural network model processes the working power global feature matrix and the space topological feature matrix through learnable neural network parameters to obtain feature expression containing irregular space topological features and working features of the terminal equipment of the Internet of things. In the topology working power global feature matrix, each row vector corresponds to a topology working power global feature vector of each internet of things terminal device.
In step S170, inter-row vector feature aggregation optimization is performed on the topology working power global feature matrix to obtain an optimized topology working power global feature matrix. Particularly, in the technical scheme of the application, when the working power global feature matrix and the spatial topology feature matrix are used for obtaining the topology working power global feature matrix through a graph neural network model, each topology working power global feature vector (for example, a row vector) of the topology working power global feature matrix expresses high-dimensional associated expression of working power associated features of a single internet of things terminal device under a spatial topology, and therefore, the problem of low feature polymerization degree may exist among the topology working power global feature vectors of the topology working power global feature matrix, which may affect the feature expression effect of the extracted classification feature vector, thereby affecting the accuracy of the classification result of the classification feature vector through a classifier.
Therefore, it is considered that if each topology working power global feature vector in the topology working power global feature vectors of the topology working power global feature matrix is regarded as a single node, the class probability feature polymerization degree between the nodes can be determined based on the topology structure of the whole node through the distance representation between the nodes, and thus, the inter-node class probability matching feature vector corresponding to each topology working power global feature vector is calculated and represented as:
wherein, the first and the second end of the pipe are connected with each other,
is each topology operating power global eigenvector of said topology operating power global eigenvector matrix,
is the topology working power global feature vector of the topology working power global feature matrix and the topology working power global feature vector
Is a distance therebetween, i.e.
Less than a predetermined threshold value (
) The topology operating power global feature vector of (a),
in order to weight the hyper-parameters,
it is shown that the difference is made by position,
representing an exponential operation of a vector, the exponential operation of the vector representing a calculation into the vectorThe characteristic value of each position is a natural exponent function value of a power.
I.e., if it is to be
As a node of the topology, then
Can be considered as being within the topology with said node
Connected nodes, i.e.
Representing nodes under a topology
And node
With edges in between. Therefore, the inter-node class probability matching feature vector is calculated to represent the interaction degree between the nodes and the adjacent nodes in the topological structure under the class probability, the topological working power global feature vector replacing the topological working power global feature matrix is rearranged into the topological working power global feature matrix, the class probability feature polymerization degree between all the nodes in the topological structure formed by multiple nodes can be improved, and the method is equivalent to applying an attention mechanism to the node features on the feature polymerization dimension based on the internal feature interaction, so that the feature polymerization degree between all the topological working power global feature vectors of the topological working power global feature matrix is improved, and the accuracy of the extracted classification result of the classification feature vector passing through the classifier is also improved.
In steps S180 and S190, extracting a row vector of the to-be-detected internet-of-things terminal device from the optimized topology working power global feature matrix as a classification feature vector, and passing the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the state of the to-be-detected internet-of-things terminal device is abnormal. The method comprises the steps of extracting a topological working characteristic expression of the terminal equipment of the Internet of things to be detected, and determining a class probability label to which the topological working characteristic expression belongs by using a classifier.
In the classification processing of the classifier, firstly, full-connection coding is carried out on the classification characteristic vectors by using a full-connection layer of the classifier to obtain coded classification characteristic vectors; then, the code classification feature vector is input into a Softmax classification function of the classifier to obtain the classification result, that is, the Softmax classification function is used to classify the code classification feature vector to obtain a first probability value that the code classification feature vector is attributed to the to-be-detected internet of things terminal device and has an abnormality (a first label) in the state, and a second probability value that the code classification feature vector is attributed to the to-be-detected internet of things terminal device and has no abnormality (a second label), and then a label corresponding to the larger one of the first probability value and the second probability value is determined as the classification result.
Fig. 3 is a flowchart of obtaining a classification result by passing the classification feature vector through a classifier in the data processing method based on the internet of things according to the embodiment of the present application. As shown in fig. 3, the step of passing the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the state of the to-be-detected internet-of-things terminal device is abnormal or not, includes the steps of: s210, carrying out full-connection coding on the classified feature vectors by using a full-connection layer of the classifier to obtain coded classified feature vectors; and S220, inputting the encoding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
In summary, the data processing method based on the internet of things technology is clarified, and the internet of things technology is utilized to directly collect relevant data of the monitored internet of things terminal equipment to perform performance monitoring and fault early warning on the monitored internet of things terminal equipment. Specifically, performing one-dimensional convolution coding and full-connection coding on the working power values of the internet of things terminal devices at a plurality of preset time points in a preset time period to extract high-dimensional implicit characteristics of the working power of the internet of things terminal devices at the time points and high-dimensional implicit association mode characteristics associated with the working power of the internet of things terminal devices at the time points; and high-dimensional local implicit associated features of the terminal equipment of the internet of things on the spatial topological dimension are mined to optimize the feature expression of the working features of the terminal equipment of the internet of things, and the classification accuracy is improved. Therefore, the state of the terminal equipment of the Internet of things is detected based on the technology of the Internet of things.
Exemplary System
Fig. 4 is a block diagram of a data processing system based on internet of things technology according to an embodiment of the application. As shown in fig. 4, a data processing system 100 based on internet of things according to an embodiment of the present application includes: the device power obtaining module 110 is configured to obtain working power values of a plurality of internet of things terminal devices at a plurality of predetermined time points within a predetermined time period; the power feature extraction module 120 is configured to arrange working power values of the internet of things terminal devices at a plurality of predetermined time points in a predetermined time period into working power input vectors according to a time dimension, and then obtain a plurality of working power feature vectors through a time sequence encoder including a one-dimensional convolutional layer and a full connection layer; a spatial topology constructing module 130, configured to construct a spatial topology matrix of the multiple internet of things terminal devices, where a value of each position on a non-diagonal position of the spatial topology matrix is a distance between two corresponding internet of things terminal devices; a topological feature extraction module 140, configured to pass the spatial topological matrix through a convolutional neural network model as a feature extractor to obtain a spatial topological feature matrix; a global arrangement module 150, configured to perform two-dimensional arrangement on the plurality of working power eigenvectors to obtain a working power global eigenvector matrix; the graph neural network coding module 160 is configured to pass the working power global feature matrix and the spatial topological feature matrix through a graph neural network model to obtain a topological working power global feature matrix; the inter-vector feature polymerization degree optimization module 170 is configured to perform inter-row vector feature polymerization degree optimization on the topology working power global feature matrix to obtain an optimized topology working power global feature matrix; the extraction module 180 of the to-be-detected internet of things terminal equipment is used for extracting row vectors of the to-be-detected internet of things terminal equipment from the optimized topology working power global feature matrix as classification feature vectors; and the detection result generation module 190 is configured to pass the classification feature vectors through a classifier to obtain a classification result, where the classification result is used to indicate whether the state of the to-be-detected internet of things terminal device is abnormal.
In an example, in the data processing system 100 based on internet of things, the power feature extraction module 120 is further configured to: performing one-dimensional convolution encoding on the working power input vector by using a one-dimensional convolution layer of the time sequence encoder according to the following formula so as to extract high-dimensional implicit correlation characteristics among characteristic values of all positions in the working power input vector, wherein the formula is as follows:
wherein a is the width of the convolution kernel in the x direction,
Is a convolution kernel parameter vector,
Is a matrix of local vectors operating with a convolution kernel, w is the size of the convolution kernel,
representing the operating power input vector and,
representing a one-dimensional convolution coding operation on the working power input vector; and using a full connection layer of the time sequence encoder to perform full connection encoding on the high-dimensional implicit associated features by using the following formula so as to extract the high-dimensional implicit features of feature values of all positions in the high-dimensional implicit associated features, wherein the formula is as follows:
wherein
Is the high-dimensional implicit associative feature that is,
is the output vector of the output vector,
is a matrix of the weights that is,
is a vector of the offset to the offset,
representing a matrix multiplication.
In an example, in the data processing system 100 based on internet of things, the topological feature extraction module 140 is further configured to: using the layers of the convolutional neural network model in forward pass of the layers respectively: performing convolution processing on input data to obtain a convolution characteristic diagram; performing mean pooling based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; and the output of the last layer of the convolutional neural network model is the spatial topological characteristic matrix, and the input of the first layer of the convolutional neural network model is the spatial topological matrix.
In one example, in the data processing system 100 based on the internet of things, the convolutional neural network model as the feature extractor is a deep residual network.
In one example, in the data processing system 100 based on internet of things, the deep residual network is ResNet 150.
In an example, in the data processing system 100 based on the internet of things, the inter-vector feature aggregation degree optimizing module 170 is further configured to: performing characteristic polymerization degree optimization among row vectors on the topology working power global characteristic matrix according to the following formula to obtain an optimized topology working power global characteristic matrix; wherein the formula is:
wherein the content of the first and second substances,
is each topology operating power global eigenvector of said topology operating power global eigenvector matrix,
is the topology working power global feature vector of the topology working power global feature matrix and the topology working power global feature vector
Is a distance therebetween, i.e.
Less than a predetermined threshold value (
) The topology operating power global feature vector of (a),
in order to weight the hyper-parameters,
it is shown that the difference is made by position,
an exponential operation of a vector representing a calculation of a natural exponential function value raised to a power of a feature value of each position in the vector is represented.
In an example, in the data processing system 100 based on internet of things, the detection result generating module 190 is further configured to: performing full-concatenation coding on the classification feature vectors by using a full-concatenation layer of the classifier to obtain coded classification feature vectors; and inputting the encoding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the internet of things based data processing system 100 described above have been described in detail in the description of the internet of things based data processing method described above with reference to fig. 1 to 3, and thus, a repetitive description thereof will be omitted.
As described above, the data processing system 100 based on the internet of things according to the embodiment of the present application may be implemented in various terminal devices, for example, a server for data processing based on the internet of things. In one example, the data processing system 100 based on internet of things according to the embodiment of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the internet of things technology-based data processing system 100 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the data processing system 100 based on the internet of things technology may also be one of many hardware modules of the terminal device.
Alternatively, in another example, the data processing system 100 based on internet of things and the terminal device may be separate devices, and the data processing system 100 based on internet of things may be connected to the terminal device through a wired and/or wireless network and transmit the interaction information according to the agreed data format.
Exemplary electronic device
Next, an electronic apparatus according to an embodiment of the present application is described with reference to fig. 5. FIG. 5 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application. As shown in fig. 5, the electronic device 10 includes one or more processors 11 and memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other form of connection mechanism (not shown).
The input device 13 may include, for example, a keyboard, a mouse, and the like.
The output device 14 can output various information including the classification result to the outside. The output devices 14 may include, for example, a display, speakers, a printer, and a communication network and its connected remote output devices, among others.
Of course, for simplicity, only some of the components of the electronic device 10 relevant to the present application are shown in fig. 5, and components such as buses, input/output interfaces, and the like are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments of the present application may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the functions of the internet of things technology-based data processing method according to various embodiments of the present application described in the above-mentioned "exemplary methods" section of this specification.
The computer program product may be written with program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform steps in functions in a data processing method based on internet of things technology according to various embodiments of the present application described in the above section "exemplary method" of the present specification.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing describes the general principles of the present application in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present application are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.
The block diagrams of devices, apparatuses, systems referred to in this application are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations are to be considered as equivalents of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
Claims (10)
1. A data processing method based on the technology of the Internet of things is characterized by comprising the following steps:
the method comprises the steps of obtaining working power values of a plurality of internet of things terminal devices at a plurality of preset time points in a preset time period;
respectively arranging the working power values of a plurality of preset time points of each Internet of things terminal device in a preset time period into working power input vectors according to the time dimension, and then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer;
constructing a spatial topological matrix of the plurality of internet of things terminal devices, wherein the value of each position on the off-diagonal position of the spatial topological matrix is the distance between two corresponding internet of things terminal devices;
passing the spatial topological matrix through a convolutional neural network model serving as a feature extractor to obtain a spatial topological feature matrix;
the working power eigenvectors are arranged in a two-dimensional mode to obtain a working power global eigenvector matrix;
passing the working power global feature matrix and the spatial topological feature matrix through a graph neural network model to obtain a topological working power global feature matrix;
performing characteristic polymerization degree optimization among row vectors on the topology working power global characteristic matrix to obtain an optimized topology working power global characteristic matrix;
extracting row vectors of the terminal equipment of the Internet of things to be detected from the optimized topology working power global feature matrix to serve as classification feature vectors; and
and enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the state of the to-be-detected Internet of things terminal equipment is abnormal or not.
2. The data processing method based on the internet of things technology of claim 1, wherein the obtaining of the plurality of working power feature vectors by the sequential encoder including the one-dimensional convolutional layer and the fully-connected layer after the working power values of the plurality of predetermined time points of each terminal device of the internet of things in the predetermined time period are respectively arranged as the working power input vector according to the time dimension comprises:
performing one-dimensional convolution encoding on the working power input vector by using a one-dimensional convolution layer of the time sequence encoder according to the following formula so as to extract high-dimensional implicit correlation characteristics among characteristic values of all positions in the working power input vector, wherein the formula is as follows:
wherein a is the width of the convolution kernel in the x direction,
Is a convolution kernel parameter vector,
Is a matrix of local vectors operating with a convolution kernel, w is the size of the convolution kernel,
represents the operating power input vector and is,
representing a one-dimensional convolution coding operation on the working power input vector; and
using a full-connection layer of the time sequence encoder to perform full-connection encoding on the high-dimensional implicit associated features by using the following formula to extract the high-dimensional implicit features of feature values of all positions in the high-dimensional implicit associated features, wherein the formula is as follows:
wherein
Is the high-dimensional implicit associative feature that is,
is the output vector of the output vector,
is a matrix of the weights that is,
is a vector of the offset to the offset,
representing a matrix multiplication.
3. The data processing method based on the internet of things technology as claimed in claim 2, wherein the passing the spatial topological matrix through a convolutional neural network model as a feature extractor to obtain a spatial topological feature matrix comprises:
using the layers of the convolutional neural network model in forward pass of the layers respectively:
performing convolution processing on input data to obtain a convolution characteristic diagram;
performing mean pooling based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and
performing nonlinear activation on the pooled feature map to obtain an activated feature map;
and the output of the last layer of the convolutional neural network model is the spatial topological characteristic matrix, and the input of the first layer of the convolutional neural network model is the spatial topological matrix.
4. The data processing method based on the internet of things technology as claimed in claim 3, wherein the convolutional neural network model as the feature extractor is a deep residual network.
5. The Internet of things technology-based data processing method according to claim 4, wherein the deep residual network is ResNet 150.
6. The data processing method based on the internet of things technology of claim 5, wherein the optimizing the inter-row vector feature polymerization degree of the topology working power global feature matrix to obtain an optimized topology working power global feature matrix comprises:
performing characteristic polymerization degree optimization among row vectors on the topology working power global characteristic matrix according to the following formula to obtain an optimized topology working power global characteristic matrix;
wherein the formula is:
wherein the content of the first and second substances,
is each topology operating power global eigenvector of said topology operating power global eigenvector matrix,
is the topology working power global feature vector of the topology working power global feature matrix and the topology working power global feature vector
Is a distance therebetween, i.e.
Less than a predetermined threshold value (
) The topology operating power global feature vector of (a),
in order to weight the hyper-parameters,
it is shown that the difference is made by position,
an exponential operation of a vector representing a calculation of a natural exponential function value raised to a power of a feature value of each position in the vector is represented.
7. The data processing method based on the internet of things technology of claim 6, wherein the step of passing the classification feature vectors through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the state of the terminal device of the internet of things to be detected is abnormal or not comprises the steps of:
performing full-concatenation coding on the classification feature vectors by using a full-concatenation layer of the classifier to obtain coded classification feature vectors; and
inputting the encoding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
8. A data processing system based on Internet of things technology is characterized by comprising:
the device power acquisition module is used for acquiring the working power values of a plurality of Internet of things terminal devices at a plurality of preset time points in a preset time period;
the power characteristic extraction module is used for arranging working power values of a plurality of preset time points of each Internet of things terminal device in a preset time period into working power input vectors according to a time dimension, and then obtaining a plurality of working power characteristic vectors through a time sequence encoder comprising a one-dimensional convolution layer and a full connection layer;
the space topology construction module is used for constructing a space topology matrix of the plurality of internet of things terminal devices, and the value of each position on the off-diagonal position of the space topology matrix is the distance between two corresponding internet of things terminal devices;
the topological characteristic extraction module is used for enabling the space topological matrix to pass through a convolutional neural network model serving as a characteristic extractor so as to obtain a space topological characteristic matrix;
the global arrangement module is used for carrying out two-dimensional arrangement on the working power eigenvectors to obtain a working power global eigenvector matrix;
the graph neural network coding module is used for enabling the working power global characteristic matrix and the spatial topological characteristic matrix to pass through a graph neural network model so as to obtain a topological working power global characteristic matrix;
the inter-vector feature polymerization degree optimization module is used for performing inter-row vector feature polymerization degree optimization on the topology working power global feature matrix to obtain an optimized topology working power global feature matrix;
the extraction module of the to-be-detected Internet of things terminal equipment is used for extracting row vectors of the to-be-detected Internet of things terminal equipment from the optimized topology working power global feature matrix as classification feature vectors; and
and the detection result generation module is used for enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the state of the to-be-detected Internet of things terminal equipment is abnormal or not.
9. The internet of things technology-based data processing system of claim 8, wherein the topological feature extraction module is further configured to:
using the layers of the convolutional neural network model in forward pass of the layers respectively:
performing convolution processing on input data to obtain a convolution characteristic diagram;
performing mean pooling based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and
performing nonlinear activation on the pooled feature map to obtain an activated feature map;
and the output of the last layer of the convolutional neural network model is the spatial topological characteristic matrix, and the input of the first layer of the convolutional neural network model is the spatial topological matrix.
10. The internet-of-things-based data processing system of claim 9, wherein the convolutional neural network model as the feature extractor is a deep residual network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310128375.5A CN115834433B (en) | 2023-02-17 | 2023-02-17 | Data processing method and system based on Internet of things technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310128375.5A CN115834433B (en) | 2023-02-17 | 2023-02-17 | Data processing method and system based on Internet of things technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115834433A true CN115834433A (en) | 2023-03-21 |
| CN115834433B CN115834433B (en) | 2023-05-12 |
Family
ID=85521753
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310128375.5A Active CN115834433B (en) | 2023-02-17 | 2023-02-17 | Data processing method and system based on Internet of things technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115834433B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116095089A (en) * | 2023-04-11 | 2023-05-09 | 云南远信科技有限公司 | Remote sensing satellite data processing method and system |
| CN116247824A (en) * | 2023-03-30 | 2023-06-09 | 国网河南省电力公司安阳供电公司 | Control method and system for power equipment |
| CN117061322A (en) * | 2023-09-27 | 2023-11-14 | 广东云百科技有限公司 | Internet of things flow pool management method and system |
| CN117274903A (en) * | 2023-09-25 | 2023-12-22 | 安徽南瑞继远电网技术有限公司 | Intelligent early warning device and method for electric power inspection based on intelligent AI chip |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113222265A (en) * | 2021-05-21 | 2021-08-06 | 内蒙古大学 | Mobile multi-sensor space-time data prediction method and system in Internet of things |
| US20210406603A1 (en) * | 2020-06-26 | 2021-12-30 | Tata Consultancy Services Limited | Neural networks for handling variable-dimensional time series data |
| CN114077811A (en) * | 2022-01-19 | 2022-02-22 | 华东交通大学 | Electric power Internet of things equipment abnormality detection method based on graph neural network |
| US20220076101A1 (en) * | 2020-09-04 | 2022-03-10 | Alipay (Hangzhou) Information Technology Co., Ltd. | Object feature information acquisition, classification, and information pushing methods and apparatuses |
| US20220101556A1 (en) * | 2020-09-29 | 2022-03-31 | International Business Machines Corporation | Computer automated interactive activity recognition based on keypoint detection |
| CN114647198A (en) * | 2022-03-09 | 2022-06-21 | 深圳市经纬纵横科技有限公司 | Intelligent home control method and system based on Internet of things and electronic equipment |
| CN115434873A (en) * | 2022-08-23 | 2022-12-06 | 华能新能源股份有限公司 | Offshore wind turbine tower inclination online monitoring system and method thereof |
| CN115586757A (en) * | 2022-11-09 | 2023-01-10 | 广东金柳信息科技有限公司 | Intelligent control system and method for mechanical equipment |
-
2023
- 2023-02-17 CN CN202310128375.5A patent/CN115834433B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210406603A1 (en) * | 2020-06-26 | 2021-12-30 | Tata Consultancy Services Limited | Neural networks for handling variable-dimensional time series data |
| US20220076101A1 (en) * | 2020-09-04 | 2022-03-10 | Alipay (Hangzhou) Information Technology Co., Ltd. | Object feature information acquisition, classification, and information pushing methods and apparatuses |
| US20220101556A1 (en) * | 2020-09-29 | 2022-03-31 | International Business Machines Corporation | Computer automated interactive activity recognition based on keypoint detection |
| CN113222265A (en) * | 2021-05-21 | 2021-08-06 | 内蒙古大学 | Mobile multi-sensor space-time data prediction method and system in Internet of things |
| CN114077811A (en) * | 2022-01-19 | 2022-02-22 | 华东交通大学 | Electric power Internet of things equipment abnormality detection method based on graph neural network |
| CN114647198A (en) * | 2022-03-09 | 2022-06-21 | 深圳市经纬纵横科技有限公司 | Intelligent home control method and system based on Internet of things and electronic equipment |
| CN115434873A (en) * | 2022-08-23 | 2022-12-06 | 华能新能源股份有限公司 | Offshore wind turbine tower inclination online monitoring system and method thereof |
| CN115586757A (en) * | 2022-11-09 | 2023-01-10 | 广东金柳信息科技有限公司 | Intelligent control system and method for mechanical equipment |
Non-Patent Citations (2)
| Title |
|---|
| 戚琦;申润业;王敬宇;: "GAD：基于拓扑感知的时间序列异常检测", 通信学报 * |
| 郭嘉琰;李荣华;张岩;王国仁;: "基于图神经网络的动态网络异常检测算法", 软件学报 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116247824A (en) * | 2023-03-30 | 2023-06-09 | 国网河南省电力公司安阳供电公司 | Control method and system for power equipment |
| CN116247824B (en) * | 2023-03-30 | 2023-11-17 | 国网河南省电力公司安阳供电公司 | Control method and system for power equipment |
| CN116095089A (en) * | 2023-04-11 | 2023-05-09 | 云南远信科技有限公司 | Remote sensing satellite data processing method and system |
| CN116095089B (en) * | 2023-04-11 | 2023-06-16 | 云南远信科技有限公司 | Remote sensing satellite data processing method and system |
| CN117274903A (en) * | 2023-09-25 | 2023-12-22 | 安徽南瑞继远电网技术有限公司 | Intelligent early warning device and method for electric power inspection based on intelligent AI chip |
| CN117274903B (en) * | 2023-09-25 | 2024-04-19 | 安徽南瑞继远电网技术有限公司 | Intelligent early warning device and method for electric power inspection based on intelligent AI chip |
| CN117061322A (en) * | 2023-09-27 | 2023-11-14 | 广东云百科技有限公司 | Internet of things flow pool management method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115834433B (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115834433A (en) | Data processing method and system based on Internet of things technology | |
| US10354204B2 (en) | Machine learning predictive labeling system | |
| CN115796173B (en) | Data processing method and system for supervising reporting requirements | |
| Chen et al. | An efficient network behavior anomaly detection using a hybrid DBN-LSTM network | |
| CN115099684B (en) | Enterprise safety production management system and management method thereof | |
| CN111738520A (en) | System load prediction method fusing isolated forest and long-short term memory network | |
| CN116579618B (en) | Data processing method, device, equipment and storage medium based on risk management | |
| CN116015837A (en) | Intrusion detection method and system for computer network information security | |
| CN116257406A (en) | Gateway data management method and system for smart city | |
| CN116247824B (en) | Control method and system for power equipment | |
| Su et al. | Nonlinear compensation algorithm for multidimensional temporal data: A missing value imputation for the power grid applications | |
| CN117077075A (en) | Water quality monitoring system and method for environmental protection | |
| CN116151845A (en) | Product full life cycle management system and method based on industrial Internet of things technology | |
| CN116216234A (en) | Intelligent monitoring system and method for coal conveying system based on video | |
| CN116826961A (en) | Intelligent power grid dispatching and operation and maintenance system, method and storage medium | |
| CN116150371A (en) | Asset repayment plan mass data processing method based on sharingJDBC | |
| CN114722388A (en) | Database data information security monitoring method | |
| CN113892910A (en) | Internet of things-based intelligent closestool old people monitoring method | |
| CN113392278A (en) | Industrial internet-based equipment liquid pipeline flow detection method | |
| Shen et al. | Long-term multivariate time series forecasting in data centers based on multi-factor separation evolutionary spatial–temporal graph neural networks | |
| CN115993507B (en) | Fault distance measurement method of electrified railway traction power supply system | |
| EP4379728A1 (en) | Training method and apparatus for property model, and electronic device, computer-readable storage medium and computer program product | |
| CN116561696B (en) | Multi-dimensional user adjustable load rapid aggregation method and system thereof | |
| CN116486345A (en) | Property service platform management system and method thereof | |
| CN107918670A (en) | A kind of alert processing method applied to power communication system |
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 |