CN115510900A - Automatic circuit fault diagnosis method and system - Google Patents
Automatic circuit fault diagnosis method and system Download PDFInfo
- Publication number
- CN115510900A CN115510900A CN202211130395.8A CN202211130395A CN115510900A CN 115510900 A CN115510900 A CN 115510900A CN 202211130395 A CN202211130395 A CN 202211130395A CN 115510900 A CN115510900 A CN 115510900A
- Authority
- CN
- China
- Prior art keywords
- feature
- waveform
- classification
- neural network
- matrix
- 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.)
- Withdrawn
Links
Images
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
Disclosed are a circuit fault automatic diagnosis method and a diagnosis system thereof, which collect a plurality of sound signals through a plurality of MEMS sound sensors deployed in a laboratory in a preset topological style and construct a topological matrix of the plurality of MEMS sound sensors. Then, a waveform diagram and a topological matrix of the sound signal are coded and fused by using a deep neural network model to obtain a classification feature vector containing irregular space topological information and high-dimensional sound features, and a judgment result of whether a low-voltage arc fault exists in a circuit of a laboratory is obtained through a classifier. In this way, fault diagnosis is performed based on the deep neural network model of deep learning to improve the accuracy of laboratory circuit fault diagnosis.
Description
Technical Field
The present application relates to the field of artificial intelligence technology, and more particularly, to a method and a system for automatically diagnosing a circuit fault.
Background
The laboratory is an important place for scientific research personnel to conduct academic research and plays a very important role in scientific and technological development. In recent years, reports about fire disasters in laboratories of colleges and universities are not interrupted, which causes experimental data loss, huge economic loss and even serious threat to human life safety. Safe power utilization is one of the key factors for avoiding fire accidents in laboratories. The laboratory is with the condition that the electric wire is ageing, and the power consumption load is high or overload power consumption, all can lead to the electric arc trouble to take place, or damage laboratory glassware, or cause electric fire accident.
Therefore, a fault diagnosis and early warning scheme for laboratory circuit safety is desired.
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.
Deep learning and development of neural networks provide new solutions for fault diagnosis and early warning of laboratory circuit safety.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a circuit fault automatic diagnosis method and a circuit fault automatic diagnosis system, wherein a plurality of MEMS sound sensors deployed in a laboratory in a preset topological style are used for collecting a plurality of sound signals and constructing a topological matrix of the MEMS sound sensors. Then, a waveform diagram and a topological matrix of the sound signal are coded and fused by using a deep neural network model to obtain a classification feature vector containing irregular space topological information and high-dimensional sound features, and a judgment result of whether a low-voltage arc fault exists in a circuit of a laboratory is obtained through a classifier. In this way, fault diagnosis is performed based on the deep neural network model for deep learning to improve the accuracy of laboratory circuit fault diagnosis.
According to an aspect of the present application, there is provided a circuit fault automatic diagnosis method, including: acquiring a plurality of sound signals through a plurality of MEMS sound sensors deployed in a laboratory in a preset topological pattern; passing the waveform map of each of the acoustic signals through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps; acquiring a topological matrix of the MEMS sound sensors, wherein the value of each position on a non-diagonal position in the topological matrix is the distance between two corresponding MEMS sound sensors, and the value of each position on a diagonal in the topological matrix is zero; passing the topological matrix through a second convolutional neural network to obtain a topological feature matrix; arranging the plurality of waveform feature maps into an input tensor along a channel dimension, and then obtaining a waveform feature vector through a third convolution neural network using a three-dimensional convolution kernel; performing label value response mapping on the feature values of the positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the label value response mapping is performed based on a classification label value and a probability value of passing the waveform feature vector through a classifier with the classification label value as a classification label; multiplying the topological feature matrix with the corrected waveform feature vector to map the corrected waveform feature vector into a high-dimensional feature space of the topological feature matrix to generate a classification feature vector; and passing the classified feature vectors through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a low-voltage arc fault exists in a circuit of a laboratory.
According to another aspect of the present application, there is provided a circuit fault automatic diagnosis system including: the system comprises a sound data acquisition unit, a sound processing unit and a control unit, wherein the sound data acquisition unit is used for acquiring a plurality of sound signals through a plurality of MEMS sound sensors which are deployed in a laboratory in a preset topological style; a sound feature extraction unit, which is used for passing the waveform diagram of each sound signal through a first convolution neural network using a channel attention mechanism to obtain a plurality of waveform feature diagrams; a topology matrix obtaining unit, configured to obtain a topology matrix of the multiple MEMS acoustic sensors, where a value of each position on a non-diagonal position in the topology matrix is a distance between two corresponding MEMS acoustic sensors, and a value of each position on a diagonal position in the topology matrix is zero; the topology coding unit is used for enabling the topology matrix to pass through a second convolutional neural network so as to obtain a topology characteristic matrix; the sound correlation coding unit is used for arranging the waveform feature maps into an input tensor along the channel dimension and then obtaining a waveform feature vector by using a third convolution neural network of a three-dimensional convolution kernel; a feature vector correction unit configured to perform label value response mapping on feature values of respective positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the label value response mapping is performed based on a classification label value and a probability value of passing the waveform feature vector through a classifier having the classification label value as a classification label; the characteristic fusion unit is used for multiplying the topological characteristic matrix and the correction waveform characteristic vector and mapping the correction waveform characteristic vector to a high-dimensional characteristic space of the topological characteristic matrix to generate a classification characteristic vector; and a diagnostic result generation unit for passing the classification feature vector through a classifier to obtain a classification result, which is used for indicating whether a low-voltage arc fault exists in a circuit of a laboratory.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory in which are stored computer program instructions which, when executed by the processor, cause the processor to carry out the circuit fault automatic diagnosis 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 execute the circuit fault auto-diagnosis method as described above.
Compared with the prior art, the circuit fault automatic diagnosis method and the circuit fault automatic diagnosis system have the advantages that the multiple MEMS sound sensors deployed in a laboratory in a preset topology mode acquire multiple sound signals and construct a topology matrix of the multiple MEMS sound sensors. Then, a waveform diagram and a topological matrix of the sound signal are coded and fused by using a deep neural network model to obtain a classification feature vector containing irregular space topological information and high-dimensional sound features, and a judgment result of whether a low-voltage arc fault exists in a circuit of a laboratory is obtained through a classifier. In this way, fault diagnosis is performed based on the deep neural network model for deep learning to improve the accuracy of laboratory circuit fault diagnosis.
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 indicate like parts or steps.
Fig. 1 illustrates an application scenario diagram of a circuit fault automatic diagnosis method and a diagnosis system thereof according to an embodiment of the present application.
Fig. 2 illustrates a flow chart of a circuit fault automatic diagnostic method according to an embodiment of the present application.
Fig. 3 illustrates a schematic diagram of a system architecture of a circuit fault automatic diagnosis method according to an embodiment of the present application.
Fig. 4 is a flowchart illustrating that, in the method for automatically diagnosing a circuit fault according to the embodiment of the present application, the waveform of each of the sound signals is passed through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps.
Fig. 5 is a flowchart illustrating that the topology matrix passes through a second convolutional neural network to obtain a topology feature matrix in the automatic circuit fault diagnosis method according to the embodiment of the application.
Fig. 6 is a flowchart illustrating that the classified feature vectors are passed through a classifier to obtain a classification result in the automatic circuit fault diagnosis method according to the embodiment of the application.
FIG. 7 illustrates a block diagram schematic of a circuit fault automatic diagnostic system according to an embodiment of the present application.
Fig. 8 illustrates a block diagram of a diagnosis result generation unit in the circuit failure automatic diagnosis system according to the embodiment of the present application.
FIG. 9 illustrates a block diagram of an electronic device in accordance with 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 apparent that the described embodiments are only a few embodiments of the present application, and not all embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.
Overview of a scene
As described above, safe power utilization is one of the key factors for avoiding a fire accident in a laboratory. The laboratory is with the condition that the electric wire is ageing, and the power consumption load is high or overload power consumption, all can lead to the electric arc trouble to take place, or damage laboratory glassware, or cause electric fire accident. Deep learning and development of neural networks (especially development of machine vision technology) provide solutions and schemes for automatic circuit fault diagnosis.
In particular, through research, the inventor of the present application found that if a circuit fault, for example, a low voltage arc fault, occurs in a laboratory, a specific sound effect is generated, and thus, if sound data in the laboratory can be collected and pattern recognition is performed on the sound data, a fault pre-warning for laboratory circuit safety can be performed. Accordingly, the present inventors attempted to perform low-voltage arc fault detection using the sound characteristic as a non-electrical characteristic amount. For example, a low voltage arc fault detection may be performed by transmitting the collected acoustic signal to a neural network using a device such as a MEMS acoustic sensor. However, on the one hand, the non-electrical features are susceptible to environmental disturbances, and on the other hand, the weak strength of the non-electrical features requires feature reinforcement.
In order to solve the technical problem, in the technical solution of the present application, a plurality of MEMS acoustic sensors are deployed in a preset topological pattern in a laboratory to collect acoustic signals from a plurality of locations of the laboratory through the plurality of MEMS acoustic sensors. On one hand, the sound collection range can be expanded through the sound data of the multiple positions, and on the other hand, the sound data of the multiple positions are related, so that the mutual cross validation can be realized to improve the richness and the representation capability of the data.
Specifically, in the technical solution of the present application, considering that the sound signal is also essentially two-dimensional image data, which is different from a conventional processing means of the sound signal, in the embodiment of the present application, a convolutional neural network model is used as a feature extractor to encode a waveform map of the sound signal. Specifically, the waveform diagram of each sound signal is passed through a first convolutional neural network to obtain a plurality of waveform feature diagrams. The channel dimension of the convolutional neural network is more focused on the difference between different objects in the processed data, and therefore, in order to enhance feature extraction, in this embodiment of the present application, a channel attention mechanism is integrated into the convolutional neural network model, so that the convolutional neural network model is more focused on the difference between sound features generated by various objects (for example, noise generated by an interfering object, sound effect generated by a low-voltage arc fault, and the like) in the sound signal in the process of encoding the waveform diagram of the sound signal, so as to improve the accuracy of fault detection.
Meanwhile, in order to extract the correlation between the plurality of acoustic signals, in the embodiment of the present application, the plurality of waveform feature maps of the plurality of acoustic signals are arranged as an input tensor along the channel dimension, and then a waveform feature vector is obtained by using a third convolutional neural network of a three-dimensional convolutional kernel. That is, the feature representations of the plurality of sound signals in the high-dimensional feature space are cascaded in the data plane by channel dimension to obtain a three-dimensional input tensor, and then the three-dimensional input tensor is passed through a convolutional neural network model (third convolutional neural network) using a three-dimensional convolution kernel to extract a high-dimensional implicit expression of associated features of the sound features of the plurality of sound signals in the high-dimensional feature space to obtain a waveform feature vector.
As described above, in the embodiment of the present application, the MEMS acoustic sensors are deployed in a preset topological pattern in a laboratory, and therefore, the feature representations of the acoustic signals collected by the MEMS acoustic sensors in the high-dimensional feature space also exhibit a certain spatial topological relation. Correspondingly, in the technical solution of the present application, a convolutional neural network model (second convolutional neural network) is also used to encode a topological matrix of the plurality of MEMS acoustic sensors to obtain a topological feature matrix, where a value of each position on a non-diagonal position in the topological matrix is a distance between two corresponding MEMS acoustic sensors, and a value of each position on a diagonal position in the topological matrix is zero. And then, mapping the topological information of the topological characteristic matrix to a high-dimensional characteristic space where the waveform characteristic vector is located in a matrix multiplication mode to obtain a classified characteristic vector containing irregular space topological information and high-dimensional sound characteristics.
Particularly, in the technical solution of the present application, before the waveform feature vector is fused with the topological feature matrix, a label value response mapping is performed on the waveform feature vector V, that is:
f'＝e j2π[psin2πf+(1-p)]
j is a label value, f is a feature value of each position of the waveform feature vector V, and p is a probability value of the waveform feature vector V under the label.
Since the realized local feature attention mechanism in the channel dimension by taking the three-dimensional convolution kernel of the third convolution neural network as the filter is relative to the overall channel attention mechanism in the channel dimension of the first convolution neural network, so that the feature distribution of the waveform feature vector V has position sensitivity relative to the label value, the feature distribution is stacked as a depth structure in the solution space of the classification problem based on the feature value and the label value based on the response of the feature value position relative to the label probability, and the interpretability of the classification solution on the extraction of the model features is improved in a similar response angle mode, so that the classification effect of the waveform feature vector V is improved.
Based on this, the present application provides a circuit fault automatic diagnosis method, which includes: acquiring a plurality of sound signals through a plurality of MEMS sound sensors deployed in a laboratory in a preset topological style; passing the waveform map of each of the acoustic signals through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps; acquiring a topological matrix of the MEMS sound sensors, wherein the value of each position on a non-diagonal position in the topological matrix is the distance between two corresponding MEMS sound sensors, and the value of each position on a diagonal in the topological matrix is zero; passing the topological matrix through a second convolutional neural network to obtain a topological feature matrix; arranging the plurality of waveform feature maps into an input tensor along a channel dimension, and then obtaining a waveform feature vector through a third convolution neural network using a three-dimensional convolution kernel; performing Cauchy normalization on the feature values of the positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein tag value response mapping is performed on the feature values of the positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the tag value response mapping is performed based on a classification tag value and a probability value of passing the waveform feature vector through a classifier with the classification tag value as a classification tag; multiplying the topological feature matrix with the corrected waveform feature vector to map the corrected waveform feature vector into a high-dimensional feature space of the topological feature matrix to generate a classified feature vector; and passing the classified feature vectors through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a low-voltage arc fault exists in a circuit of a laboratory.
Fig. 1 illustrates an application scenario diagram of a circuit fault automatic diagnosis method and a diagnosis system thereof according to an embodiment of the present application.
As shown in fig. 1, in this application scenario, a plurality of MSMS sound sensors (e.g., C1 to Cn as illustrated in fig. 1) are deployed at a plurality of locations of a laboratory (e.g., B as illustrated in fig. 1) to collect a plurality of sound signals by the plurality of MSMS sound sensors. The collected plurality of sound signals and the topological matrix of the plurality of MSMS sound sensors are then input into a server (e.g., S illustrated in fig. 1) deployed with a circuit fault automatic diagnostic algorithm, wherein the server is capable of processing the plurality of sound signals and the topological matrix of the plurality of MSMS sound sensors using the circuit fault automatic diagnostic algorithm to generate a diagnostic result of whether a low voltage arc fault is present in a circuit of a laboratory.
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. 2 illustrates a flow chart of a circuit fault automatic diagnostic method according to an embodiment of the present application.
As shown in fig. 2, the method for automatically diagnosing a circuit fault according to the embodiment of the present application includes: s110, acquiring a plurality of sound signals through a plurality of MEMS sound sensors deployed in a laboratory in a preset topological style; s120, enabling the waveform diagram of each sound signal to pass through a first convolution neural network using a channel attention mechanism to obtain a plurality of waveform characteristic diagrams; s130, acquiring a topological matrix of the MEMS sound sensors, wherein the value of each position on the non-diagonal position in the topological matrix is the distance between the two corresponding MEMS sound sensors, and the value of each position on the diagonal position in the topological matrix is zero; s140, passing the topological matrix through a second convolutional neural network to obtain a topological characteristic matrix; s150, arranging the waveform feature maps into an input tensor along the channel dimension, and then obtaining a waveform feature vector through a third convolution neural network using a three-dimensional convolution kernel; s160, performing label value response mapping on the feature values of all positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the label value response mapping is performed on the basis of the classification label values and the probability values of the waveform feature vector passing through a classifier with the classification label values as classification labels; s170, multiplying the topological feature matrix and the corrected waveform feature vector, and mapping the corrected waveform feature vector to a high-dimensional feature space of the topological feature matrix to generate a classified feature vector; and S180, passing the classified feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a low-voltage arc fault exists in a circuit of a laboratory or not.
Fig. 3 illustrates a schematic diagram of a system architecture of a circuit fault automatic diagnosis method according to an embodiment of the present application. In the system architecture of the embodiments of the present application, a plurality of acoustic signals are first obtained by a plurality of MEMS acoustic sensors deployed in a laboratory. Then, the waveform map of each of the sound signals is passed through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps. Then, the plurality of waveform feature maps are arranged as input tensors along the channel dimension, and a third convolutional neural network of a three-dimensional convolutional kernel is used to obtain a waveform feature vector. And then, performing label value response mapping on the characteristic values of all the positions in the waveform characteristic vector to obtain a corrected waveform characteristic vector. Simultaneously, a topological matrix of the plurality of MEMS acoustic sensors is obtained. And then passing the topological matrix through a second convolutional neural network to obtain a topological characteristic matrix. Then, multiplying the topological feature matrix and the correction waveform feature vector, and mapping the correction waveform feature vector to a high-dimensional feature space of the topological feature matrix to generate a classification feature vector; finally, the classified feature vectors are passed through a classifier to obtain a classification result, which is used to indicate whether a low-voltage arc fault exists in a circuit of a laboratory.
In step S110, a plurality of sound signals are acquired by a plurality of MEMS sound sensors deployed in a preset topological pattern in a laboratory. It should be understood that when a circuit fault occurs in a laboratory, a specific sound effect, such as arc discharge heating air, is generated, the specific sound effect has a specific tone, whether a low-voltage arc fault exists in the circuit of the laboratory can be judged through collecting sound and analyzing, meanwhile, since the sound is propagated in the air, associated sound data also occur in adjacent positions, but due to energy loss in the propagation process, the association becomes unobvious along with the increase of the distance. Meanwhile, the specific sound effect is weak and tends to be hidden due to the flow of various instruments and various personnel in the laboratory.
Therefore, in the technical scheme of the application, a plurality of MEMS sound sensors are deployed in a preset topological style in a laboratory to collect sound signals from a plurality of positions of the laboratory through the MEMS sound sensors, and signals acquired through the plurality of positions can be verified in a mutual crossing mode through the sound data of the plurality of positions to expand the sound collection range and through the association among the sound data of the plurality of positions to improve the richness and the representation capability of the data.
In step S120, the waveform map of each of the sound signals is passed through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps. It should be understood that the acquired waveform image may include sound features of a plurality of objects, when the waveform image is subjected to feature extraction by using a neural network, the features of the plurality of objects may be extracted, and in order to obtain the features of the desired object of interest, the waveform image of each of the sound signals needs to be subjected to feature extraction and noise reduction by using a first convolution neural network of a channel attention mechanism to obtain a plurality of waveform feature images of the object of interest.
Because the first convolution neural network using the channel attention mechanism focuses more on the difference between the sound characteristics generated by each object (e.g., noise generated by an interfering object, sound effect generated by a low-voltage arc fault, etc.) in the sound signal during the process of encoding the waveform of the sound signal, that is, the first convolution neural network using the channel attention mechanism extracts the waveform characteristic diagram of the object of interest, thereby improving the accuracy of fault detection.
Specifically, in the embodiment of the present application, fig. 4 illustrates a flowchart of passing a waveform diagram of each of the sound signals through a first convolution neural network using a channel attention mechanism to obtain a plurality of waveform feature diagrams in a circuit fault automatic diagnosis method according to the embodiment of the present application. As shown in fig. 4, includes: each layer of the convolutional neural network respectively performs the following operations on input data in the forward transmission of the layer: s210, performing convolution processing on the input data based on a two-dimensional convolution kernel to generate a convolution characteristic diagram; s220, performing pooling processing on the convolution feature map to generate a pooled feature map; s230, activating the pooled feature map to generate an activated feature map; s240, calculating a quotient of the feature value mean value of the feature matrix corresponding to each channel in the activation feature map and the sum of the feature value mean values of the feature matrices corresponding to all channels as a weighting coefficient of the feature matrix corresponding to each channel; s250, weighting the feature matrix of each channel by using the weighting coefficient of each channel in the activated feature map to generate a channel attention feature map; wherein the output of the last layer of the first convolutional neural network is the waveform feature map.
More specifically, in one embodiment, a convolution kernel is slid on the oscillogram, and a value is calculated at each position to extract a high-dimensional local implicit feature of the oscillogram, so as to obtain the convolution feature map; and then performing average value pooling or maximum value pooling on the convolution feature map based on a local feature matrix to obtain the pooled feature map, and extracting main features, reducing the number of parameters and reducing overfitting through global pooling. And then, selecting an activation function to activate the rows of the pooled feature maps to obtain an activation feature map, such as a Sigmoid activation function, and introducing a nonlinear factor through the activation function to increase the characterization capability of the whole network.
Then, considering that different channels of the feature map play different roles in specifying objects, that is, feature matrices of different channels have different "weights" for reflecting features of an object to be focused, a quotient of a feature value mean of a feature matrix corresponding to each channel in the activation feature map and a sum of feature value means of feature matrices corresponding to all channels is calculated as a weighting coefficient of the feature matrix corresponding to each channel, that is, the "weight" occupied by the different channels is obtained, and then the feature matrices of each channel are weighted by the weighting coefficient of each channel to generate a channel attention feature map, that is, the feature matrix of each channel is multiplied by the "weight" occupied by the different channels and weighted, thereby obtaining the channel attention feature map. In the technical solution of the present application, the channel attention feature map focuses more on differences between sound features generated by respective objects in the sound signal.
In one embodiment of the present application, the first convolutional neural network using a channel attention mechanism focuses more on the difference in timbre produced by different objects and focuses more on the sound features of a particular timbre.
In step S130, a topology matrix of the MEMS acoustic sensors is obtained, where a value of each position on a non-diagonal position in the topology matrix is a distance between two corresponding MEMS acoustic sensors, and a value of each position on a diagonal position in the topology matrix is zero. It will be appreciated that sound will travel through air, but energy losses will occur during travel. That is, as the distance between the MEMS acoustic sensor and the acoustic source increases, the weaker the acoustic signal received by the MEMS acoustic sensor is, and the stronger the acoustic signal is otherwise; that is, the farther the MEMS acoustic sensors are from each other, the less the correlation of the received acoustic signals, and vice versa. In other words, the sound signals received by the MEMS sound sensors have spatially correlated features, and in order to introduce spatial information in subsequent calculations to increase the characterization capability of the classification vector, a topological matrix needs to be constructed, where the value of each position on the non-diagonal position in the topological matrix is the distance between two corresponding MEMS sound sensors, and the value of each position on the diagonal position in the topological matrix is zero. .
In step S140, the topological matrix is passed through a second convolutional neural network to obtain a topological feature matrix. It should be understood that the plurality of MEMS acoustic sensors are deployed in a preset topological pattern in a laboratory, and therefore, the feature representations of the plurality of acoustic signals collected by the plurality of MEMS acoustic sensors in the high-dimensional feature space also exhibit a certain spatial topological relation. Namely, the topological matrix is input into a second convolutional neural network, and spatial implicit correlation characteristics of the MEMS sound sensors are extracted, so that the topological characteristic matrix is obtained.
Specifically, in the embodiment of the present application, fig. 5 illustrates a flowchart of passing the topology matrix through a second convolutional neural network to obtain a topology feature matrix in the automatic circuit fault diagnosis method according to the embodiment of the present application. As shown in fig. 5, includes: each layer of the second convolutional neural network performs the following operations on input data in forward transmission of the layer: s310, performing convolution processing based on a two-dimensional convolution kernel on the input data by using convolution units of each layer of the second convolution neural network to obtain a convolution characteristic diagram; s320, performing global mean pooling along channel dimensions on the convolution feature map by using pooling units of each layer of the second convolution neural network to obtain a pooled feature map; s330, performing nonlinear activation on the feature values of all positions in the pooled feature map by using the activation units of all layers of the second convolutional neural network to obtain an activated feature map; wherein the output of the last layer of the second convolutional neural network is the topological feature matrix.
In step S150, the plurality of waveform feature maps are arranged as an input tensor along the channel dimension and then passed through a third convolution neural network using a three-dimensional convolution kernel to obtain a waveform feature vector. It should be understood that in order to extract the correlation between the plurality of sound signals, it is necessary to arrange the plurality of waveform feature maps of the plurality of sound signals along the channel dimension as an input tensor and then obtain a waveform feature vector by a third convolution neural network using a three-dimensional convolution kernel as a filter. That is, the feature representations of the plurality of sound signals in the high-dimensional feature space are cascaded in the data plane in the channel dimension to obtain a three-dimensional input tensor, and then the three-dimensional input tensor is passed through a convolutional neural network model (third convolutional neural network) using a three-dimensional convolution kernel to extract the high-dimensional implicit expression of the associated features of the sound features of the plurality of sound signals in the high-dimensional feature space to obtain a waveform feature vector, that is, the associated feature information distributed in the space of the plurality of sound signals can be extracted by using the three-dimensional convolution kernel.
In an embodiment of the present application, arranging the plurality of waveform feature maps as input tensors along a channel dimension, and obtaining a waveform feature vector by using a third convolutional neural network of a three-dimensional convolutional kernel, includes:
the third convolutional neural network using the three-dimensional convolutional kernel processes the input tensor by the following formula to obtain the waveform eigenvector;
wherein the formula is:
wherein
Representing the feature vector of the waveform, H j 、W j And R j Respectively representing the length, width and height of the three-dimensional convolution kernel, m represents the number of (l-1) th layer characteristic diagrams,
is the convolution kernel connected to the mth feature map of the (l-1) layer, b lj For bias, f (-) represents the activation function.
In step S160, cauchy normalization is performed on the feature values of the respective positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein tag value response mapping is performed on the feature values of the respective positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the tag value response mapping is performed based on a classification tag value and a probability value of passing the waveform feature vector through a classifier having the classification tag value as a classification tag. It should be understood that, the relevant features of the waveform feature vector in the high-dimensional feature space and the relevant topological feature information of the topological feature vector are fused and a classifier is used to determine whether a low-voltage arc fault exists in a circuit of a laboratory, for example, the topological information of the topological feature matrix is mapped into the high-dimensional feature space in which the waveform feature vector exists in a matrix multiplication manner to obtain a classified feature vector containing irregular spatial topological information and high-dimensional sound features. Preferably, in the technical solution of the present application, considering that the waveform of each of the sound signals passes through a first convolutional neural network and a third convolutional neural network to obtain a waveform feature vector, the first convolutional neural network adopts an overall channel attention mechanism in a channel dimension, and the third convolutional neural network adopts a local feature attention mechanism in the channel dimension, so that before the waveform feature vector and the topological feature vector are fused, tag value response mapping is performed on the waveform feature vector.
More specifically, the waveform feature vector V is subjected to tag value response mapping, that is:
f'＝e j2π[psin2πf+(1-p)]
j is a label value, f is a feature value of each position of the waveform feature vector V, and p is a probability value of the waveform feature vector V under the label.
Since the local feature attention mechanism in the channel dimension implemented by using the three-dimensional convolution kernel of the third convolutional neural network as a filter is relative to the overall channel attention mechanism in the channel dimension of the first convolutional neural network, so that the feature distribution of the waveform feature vector V has position sensitivity relative to the tag value, the feature distribution is stacked as a depth structure in a solution space of a classification problem based on the feature value and the tag value based on the response of the feature value position relative to the tag probability, and the interpretability of the classification solution on the extraction of the model feature is improved in the form of a class response angle, so that the classification effect of the waveform feature vector V is improved.
In step S170, the topological feature matrix is multiplied by the corrected waveform feature vector to map the corrected waveform feature vector into a high-dimensional feature space of the topological feature matrix to generate a classification feature vector. It should be understood that, in order to introduce the spatial correlation features of each sound signal and perform feature fusion in a high-dimensional space, so as to enhance the characterization capability of a classification vector, it is necessary to map the topological information of the topological feature matrix into the high-dimensional feature space in which the waveform feature vector is located in a matrix multiplication manner to obtain a classification feature vector containing irregular spatial topological information and high-dimensional sound features.
In step S180, the classified feature vector is passed through a classifier to obtain a classification result, which is used to indicate whether there is a low voltage arc fault in the electrical circuit of the laboratory.
Specifically, in the embodiment of the present application, fig. 6 illustrates a flowchart of passing the classification feature vector through a classifier to obtain a classification result in the circuit fault automatic diagnosis method according to the embodiment of the present application. S410, inputting the classification feature vector into a Softmax classification function of the classifier to obtain probability values of the classification feature vector belonging to each classification label, wherein the classification labels comprise the existence of a low-voltage arc fault in a circuit of a laboratory and the absence of the low-voltage arc fault in the circuit of the laboratory; s420, determining the classification label corresponding to the maximum probability value as the classification result; the classification result is used to indicate whether a low voltage arc fault exists in the laboratory's electrical circuit.
More specifically, in the implementation of the present application, the Softmax classification function is expressed as:
in summary, the method for automatically diagnosing circuit faults according to the embodiment of the present application is illustrated, a plurality of MEMS acoustic sensors deployed in a laboratory in a preset topology pattern are used to acquire a plurality of acoustic signals and a topology matrix of the plurality of MEMS acoustic sensors, a deep neural network model is used to encode a waveform diagram and the topology matrix of the acoustic signals to obtain a classification feature vector including irregular spatial topology information and high-dimensional acoustic features, and a classifier is used to obtain a determination result whether a low-voltage arc fault exists in a circuit in the laboratory. And the feature distribution of the waveform feature vector V has position sensitivity relative to the label value by performing label value response mapping on the waveform feature vector, so that the feature distribution is stacked into a depth structure in a solution space of a classification problem based on the feature value and the label value based on the response of the feature value position relative to the label probability, thereby improving the interpretability of the classification solution on the extraction of the model features in a response angle-like mode and improving the classification effect of the waveform feature vector V.
Exemplary System
FIG. 7 illustrates a block diagram schematic of a circuit fault auto-diagnostic system according to an embodiment of the present application
As shown in fig. 7, according to the circuit fault automatic diagnosis system 700 of the embodiment of the present application, the acoustic data acquisition unit 710 is configured to acquire a plurality of acoustic signals through a plurality of MEMS acoustic sensors deployed in a laboratory in a preset topological pattern; a sound feature extraction unit 720, configured to pass the waveform map of each of the sound signals through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps; a topology matrix obtaining unit 730, configured to obtain a topology matrix of the multiple MEMS acoustic sensors, where a value of each position on a non-diagonal position in the topology matrix is a distance between two corresponding MEMS acoustic sensors, and a value of each position on a diagonal position in the topology matrix is zero; a topology coding unit 740, configured to pass the topology matrix through a second convolutional neural network to obtain a topology feature matrix; a sound correlation encoding unit 750 configured to arrange the plurality of waveform feature maps into an input tensor along a channel dimension and obtain a waveform feature vector by using a third convolutional neural network of a three-dimensional convolutional kernel; a feature vector correction unit 760 for performing a tag value response mapping on feature values of respective positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the tag value response mapping is performed based on a classification tag value and a probability value of passing the waveform feature vector through a classifier having the classification tag value as a classification tag; a feature fusion unit 770, configured to multiply the topological feature matrix with the corrected waveform feature vector, and map the corrected waveform feature vector into a high-dimensional feature space of the topological feature matrix to generate a classified feature vector; and a diagnosis result generation unit 780 for passing the classification feature vector through a classifier to obtain a classification result indicating whether there is a low-voltage arc fault in a circuit of a laboratory
In one example, in the above automatic circuit fault diagnosis system 700, the eigenvector correction unit is further configured to: performing label value response mapping on the characteristic value of each position in the waveform characteristic vector by using the following formula to obtain a corrected waveform characteristic vector;
wherein the formula is:
f'＝e j2π[psin2πf+(1-p)]
wherein j is a classification label value, f is a feature value of each position in the waveform feature vector, and p is a probability value obtained by inputting the waveform feature vector into a classifier using the classification label value as a classification label.
In an example, in the above circuit fault automatic diagnosis system 700, the topology coding unit is further configured to: performing, using the layers of the second convolutional neural network, in forward pass of layers, input data separately:
performing convolution processing based on a two-dimensional convolution kernel on the input data by using convolution units of each layer of the second convolution neural network to obtain a convolution feature map;
performing global mean pooling along channel dimensions on the convolutional feature maps by using pooling units of each layer of the second convolutional neural network to obtain pooled feature maps; and
performing nonlinear activation on the feature values of all positions in the pooled feature map by using the activation units of all layers of the second convolutional neural network to obtain an activation feature map;
wherein the output of the last layer of the second convolutional neural network is the topological feature matrix.
In one example, in the above-mentioned circuit fault automatic diagnosis system 700, as shown in fig. 8, the diagnosis result generating unit 780 includes: a soft maximum value calculation operator unit 781, configured to input the classification feature vector into a Softmax classification function of the classifier to obtain a probability value that the classification feature vector belongs to each classification tag, where the classification tag includes a low-voltage arc fault existing in a circuit of a laboratory and a low-voltage arc fault not existing in the circuit of the laboratory; and a result determining subunit 782, configured to determine the classification label corresponding to the largest one of the probability values as 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 above-described circuit malfunction automatic diagnosis system 700 have been described in detail in the above description of the circuit malfunction automatic diagnosis method with reference to fig. 1 to 6, and thus, a repetitive description thereof will be omitted.
As described above, the circuit failure automatic diagnosis system 700 according to the embodiment of the present application can be implemented in various terminal devices, such as a server or the like having the circuit failure automatic diagnosis system. In one example, the circuit fault automatic diagnosis system 700 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 circuit failure automatic diagnosis system 700 may be a software module in an operating system of the terminal device, or may be an application program developed for the terminal device; of course, the circuit fault automatic diagnosis system 700 can also be one of many hardware modules of the terminal device.
Alternatively, in another example, the circuit malfunction automatic diagnosis system 700 and the terminal device may also be separate devices, and the circuit malfunction automatic diagnosis system 700 may be connected to the terminal device through a wired and/or wireless network and transmit the mutual information according to an agreed data format.
Exemplary electronic device
Next, an electronic apparatus according to an embodiment of the present application is described with reference to fig. 9.
FIG. 9 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application.
As shown in fig. 9, 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. 9, 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 of the circuit fault auto-diagnosis method according to various embodiments of the present application described in the "exemplary methods" section of this specification above.
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 the steps of the method for automatic diagnosis of circuit failure according to various embodiments of the present application described in the "exemplary methods" section above in this 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 provided for purposes of illustration and understanding only, and is not intended to limit the application to the details which are set forth in order to provide a thorough understanding of the present application.
The block diagrams of devices, apparatuses, devices, systems referred to in this application are only used as illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured 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, each component or step can be decomposed and/or re-combined. 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 method for automatically diagnosing a circuit fault, comprising: acquiring a plurality of sound signals through a plurality of MEMS sound sensors deployed in a laboratory in a preset topological pattern; passing the waveform map of each of the acoustic signals through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps; acquiring a topological matrix of the MEMS sound sensors, wherein the value of each position on a non-diagonal position in the topological matrix is the distance between two corresponding MEMS sound sensors, and the value of each position on a diagonal in the topological matrix is zero; passing the topological matrix through a second convolutional neural network to obtain a topological feature matrix; arranging the plurality of waveform feature maps into an input tensor along a channel dimension, and then obtaining a waveform feature vector through a third convolution neural network using a three-dimensional convolution kernel; performing label value response mapping on the feature values of the positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the label value response mapping is performed based on a classification label value and a probability value of passing the waveform feature vector through a classifier with the classification label value as a classification label; multiplying the topological feature matrix with the corrected waveform feature vector to map the corrected waveform feature vector into a high-dimensional feature space of the topological feature matrix to generate a classification feature vector; and passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a low-voltage arc fault exists in a circuit of a laboratory or not.
2. The circuit fault automatic diagnosis method according to claim 1, wherein passing the waveform map of each of the sound signals through a first convolutional neural network using a channel attention mechanism to obtain a plurality of waveform feature maps includes: each layer of the convolutional neural network respectively performs the following operations on input data in the forward transmission of the layer: performing convolution processing on the input data based on a two-dimensional convolution kernel to generate a convolution feature map; pooling the convolved feature map to generate a pooled feature map; performing activation processing on the pooled feature map to generate an activated feature map; calculating the quotient of the eigenvalue mean of the eigenvalue matrix corresponding to each channel in the activation characteristic diagram and the sum of the eigenvalue mean of the eigenvalue matrix corresponding to all channels as the weighting coefficient of the eigenvalue matrix corresponding to each channel; weighting the feature matrix of each channel by using the weighting coefficient of each channel in the activation feature map to generate a channel attention feature map; wherein the output of the last layer of the first convolutional neural network is the waveform feature map.
3. The circuit fault automatic diagnosis method according to claim 2, wherein the arranging the plurality of waveform feature maps as the input tensor along the channel dimension to obtain the waveform feature vector by a third convolution neural network using a three-dimensional convolution kernel includes: the third convolutional neural network using the three-dimensional convolutional kernel processes the input tensor by the following formula to obtain the waveform eigenvector; it is composed ofWherein the formula is:
wherein
Representing the feature vector of the waveform, H j 、W j And R j Respectively represents the length, width and height of the three-dimensional convolution kernel, m represents the number of the (l-1) th layer characteristic diagram,
is the convolution kernel connected to the mth feature map of the (l-1) layer, b lj For biasing, f (-) represents the activation function.
4. The circuit fault automatic diagnosis method according to claim 3, wherein performing Cauchy normalization on the eigenvalues of the respective positions in the waveform eigenvector to obtain a corrected waveform eigenvector, comprises: performing label value response mapping on the characteristic value of each position in the waveform characteristic vector by using the following formula to obtain a corrected waveform characteristic vector; wherein the formula is: f' = e j2π[psin2πf+(1-p)] Wherein j is a classification label value, f is a feature value of each position in the waveform feature vector, and p is a probability value obtained by inputting the waveform feature vector by using the classification label value as a classifier of a classification label.
5. The circuit fault automatic diagnostic method of claim 4, wherein passing the topology matrix through a second convolutional neural network to obtain a topology feature matrix comprises: each layer of the second convolutional neural network performs the following operations on input data in forward transmission of the layer: performing convolution processing based on a two-dimensional convolution kernel on the input data by using convolution units of each layer of the second convolution neural network to obtain a convolution feature map; performing global mean pooling along channel dimensions on the convolution feature map by using pooling units of each layer of the second convolution neural network to obtain a pooled feature map; and using the activation units of each layer of the second convolutional neural network to carry out nonlinear activation on the feature values of each position in the pooled feature map so as to obtain an activated feature map; wherein the output of the last layer of the second convolutional neural network is the topological feature matrix.
6. The method of claim 5, wherein passing the classified feature vector through a classifier to obtain a classification result, the classification result being indicative of whether a low voltage arc fault exists in a laboratory circuit, comprises: inputting the classification feature vector into a Softmax classification function of the classifier to obtain probability values that the classification feature vector belongs to classification labels, wherein the classification labels comprise the presence of a low-voltage arc fault in a circuit of a laboratory and the absence of a low-voltage arc fault in a circuit of the laboratory; and determining the classification label corresponding to the maximum probability value as the classification result.
7. An automatic circuit fault diagnosis system, comprising: the system comprises a sound data acquisition unit, a sound processing unit and a control unit, wherein the sound data acquisition unit is used for acquiring a plurality of sound signals through a plurality of MEMS sound sensors which are deployed in a laboratory in a preset topological style; a sound feature extraction unit, which is used for passing the waveform diagram of each sound signal through a first convolution neural network using a channel attention mechanism to obtain a plurality of waveform feature diagrams; a topology matrix obtaining unit, configured to obtain a topology matrix of the multiple MEMS acoustic sensors, where a value of each position on a non-diagonal position in the topology matrix is a distance between two corresponding MEMS acoustic sensors, and a value of each position on a diagonal position in the topology matrix is zero; the topological coding unit is used for enabling the topological matrix to pass through a second convolutional neural network so as to obtain a topological characteristic matrix; the sound correlation coding unit is used for arranging the waveform feature maps into an input tensor along the channel dimension and then obtaining a waveform feature vector through a third convolution neural network using a three-dimensional convolution kernel; a feature vector correction unit configured to perform label value response mapping on feature values of respective positions in the waveform feature vector to obtain a corrected waveform feature vector, wherein the label value response mapping is performed based on a classification label value and a probability value of passing the waveform feature vector through a classifier having the classification label value as a classification label; the characteristic fusion unit is used for multiplying the topological characteristic matrix and the correction waveform characteristic vector to map the correction waveform characteristic vector to a high-dimensional characteristic space of the topological characteristic matrix so as to generate a classification characteristic vector; and a diagnostic result generation unit for passing the classification feature vector through a classifier to obtain a classification result, which is used for indicating whether a low-voltage arc fault exists in a circuit of a laboratory.
8. The circuit fault automatic diagnostic system of claim 7, wherein the eigenvector correction unit is further configured to: performing label value response mapping on the feature values of all positions in the waveform feature vector by using the following formula to obtain a corrected waveform feature vector; wherein the formula is:
f'＝e j2π[psin2πf+(1-p)]
wherein j is a classification label value, f is a feature value of each position in the waveform feature vector, and p is a probability value obtained by inputting the waveform feature vector into a classifier using the classification label value as a classification label.
9. The circuit fault automatic diagnostic system of claim 8, wherein the topology encoding unit is further configured to: performing, using the layers of the second convolutional neural network, in forward pass of layers, input data separately: performing convolution processing based on a two-dimensional convolution kernel on the input data by using convolution units of each layer of the second convolution neural network to obtain a convolution feature map; performing global mean pooling along channel dimensions on the convolution feature map by using pooling units of each layer of the second convolution neural network to obtain a pooled feature map; and using the activation units of each layer of the second convolutional neural network to carry out nonlinear activation on the feature values of each position in the pooled feature map so as to obtain an activated feature map; wherein the output of the last layer of the second convolutional neural network is the topological feature matrix.
10. The circuit fault automatic diagnosis system according to claim 9, wherein the diagnosis result generating unit includes: the soft maximum value operator unit is used for inputting the classification characteristic vector into a Softmax classification function of the classifier so as to obtain a probability value of the classification characteristic vector belonging to each classification label, and the classification labels comprise that a low-voltage arc fault exists in a circuit of a laboratory and a low-voltage arc fault does not exist in the circuit of the laboratory; and a result determining subunit, configured to determine the classification label corresponding to the largest one of the probability values as the classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211130395.8A CN115510900A (en) | 2022-09-16 | 2022-09-16 | Automatic circuit fault diagnosis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211130395.8A CN115510900A (en) | 2022-09-16 | 2022-09-16 | Automatic circuit fault diagnosis method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115510900A true CN115510900A (en) | 2022-12-23 |
Family
ID=84504470
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211130395.8A Withdrawn CN115510900A (en) | 2022-09-16 | 2022-09-16 | Automatic circuit fault diagnosis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115510900A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115841644A (en) * | 2022-12-29 | 2023-03-24 | 杭州毓贞智能科技有限公司 | Control system and method for urban infrastructure engineering equipment based on Internet of things |
-
2022
- 2022-09-16 CN CN202211130395.8A patent/CN115510900A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115841644A (en) * | 2022-12-29 | 2023-03-24 | 杭州毓贞智能科技有限公司 | Control system and method for urban infrastructure engineering equipment based on Internet of things |
| CN115841644B (en) * | 2022-12-29 | 2023-12-22 | 吕梁市经开区信息化投资建设有限公司 | Control system and method for urban infrastructure engineering equipment based on Internet of Things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114825257B (en) | Leakage protection device for LED lamp detection and leakage protection method thereof | |
| CN115471216B (en) | Data management method of intelligent laboratory management platform | |
| CN115783923A (en) | Elevator fault mode identification system based on big data | |
| CN114548246A (en) | Control method and system of intelligent device and electronic device | |
| CN116015837A (en) | Intrusion detection method and system for computer network information security | |
| CN116247824A (en) | Control method and system for power equipment | |
| JP7005045B2 (en) | Limit attack method against Naive Bayes classifier | |
| CN116343301B (en) | Personnel information intelligent verification system based on face recognition | |
| CN112733785A (en) | Stability detection method of information equipment based on layer depth and receptive field | |
| CN115146676A (en) | Circuit fault detection method and system | |
| CN115510900A (en) | Automatic circuit fault diagnosis method and system | |
| CN117269644A (en) | Line fault monitoring system and method for current transformer | |
| CN111028940A (en) | Multi-scale lung nodule detection method, device, equipment and medium | |
| CN112800217A (en) | Vector relevance matrix-based intelligent assessment method for government affair transaction processing report | |
| CN117390559A (en) | Urban garden monitoring method and device based on Internet of things and electronic equipment | |
| CN112960213A (en) | Intelligent package quality detection method using characteristic probability distribution representation | |
| CN117421655A (en) | Industrial Internet data stream anomaly detection method and system | |
| CN112418353A (en) | Neural network training method for battery diaphragm abnormity detection | |
| CN116698283A (en) | Intelligent monitoring system and method for load gravity center deviation | |
| CN115239215B (en) | Enterprise risk identification method and system based on deep anomaly detection | |
| CN115456176B (en) | Text matching method and system based on knowledge enhancement | |
| CN115620342A (en) | Cross-modal pedestrian re-identification method, system and computer | |
| CN115331081A (en) | Image target detection method and device | |
| CN113392278A (en) | Industrial internet-based equipment liquid pipeline flow detection method | |
| CN112489044A (en) | Graphene electrothermal film contact detection method for heating device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20221223 |