CN116908518A - Acquisition terminal with residual current detection function - Google Patents
Acquisition terminal with residual current detection function Download PDFInfo
- Publication number
- CN116908518A CN116908518A CN202310864568.7A CN202310864568A CN116908518A CN 116908518 A CN116908518 A CN 116908518A CN 202310864568 A CN202310864568 A CN 202310864568A CN 116908518 A CN116908518 A CN 116908518A
- Authority
- CN
- China
- Prior art keywords
- residual current
- current waveform
- characteristic vector
- waveform
- actual load
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 61
- 238000012937 correction Methods 0.000 claims abstract description 31
- 239000013598 vector Substances 0.000 claims description 221
- 230000004927 fusion Effects 0.000 claims description 65
- 230000003993 interaction Effects 0.000 claims description 47
- 230000009467 reduction Effects 0.000 claims description 35
- 230000002452 interceptive effect Effects 0.000 claims description 24
- 238000010586 diagram Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 20
- 230000002829 reductive effect Effects 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 7
- 238000003062 neural network model Methods 0.000 claims description 6
- 230000014759 maintenance of location Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 abstract description 10
- 238000000034 method Methods 0.000 description 33
- 238000004422 calculation algorithm Methods 0.000 description 11
- 238000013527 convolutional neural network Methods 0.000 description 11
- 238000009826 distribution Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 230000000694 effects Effects 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000011176 pooling Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 2
- 239000004020 conductor Substances 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000010183 spectrum analysis Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/0092—Arrangements for measuring currents or voltages or for indicating presence or sign thereof measuring current only
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/02—Measuring effective values, i.e. root-mean-square values
Abstract
The application discloses an acquisition terminal with residual current detection, which is used for realizing intelligent calculation of an effective value of residual current based on actual load current and detected residual current, improving the accuracy of the effective value of the residual current, optimizing a residual current detection scheme and reducing the dependence on a manually specified correction parameter set.
Description
Technical Field
The application relates to the field of intelligent detection, in particular to an acquisition terminal with residual current detection.
Background
The residual current is often caused by faults on the electricity utilization side, such as damage of an insulating layer caused by aging of a circuit or reduction of insulation of a live conductor to the ground caused by irregular construction and installation, artificial damage and the like, when the current passes through a human body from the conductor, so that a part of current in the main loop does not flow back to the main loop. Excess current beyond the safe range may lead to dangerous situations such as electric shock, fire, etc. The detection of the residual current is therefore of considerable importance.
In patent CN113721070a method of detecting the residual current value is provided, but the correction parameters in this method depend on a look-up table operation in a memory. The correction parameter set is selected based on manual regulation, and if the correction parameter set has an inaccurate or imperfect problem, the detection of the residual current has a great adverse effect. Therefore, an optimized residual current detection scheme is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an acquisition terminal with residual current detection, which is used for realizing intelligent calculation of an effective value of residual current based on actual load current and detected residual current, improving the accuracy of the effective value of the residual current, optimizing a residual current detection scheme and reducing the dependence on a manually specified correction parameter set.
According to one aspect of the present application, there is provided an acquisition terminal with residual current detection, comprising:
the current signal acquisition module is used for acquiring an actual load current signal and detecting a residual current signal;
the current characteristic interaction module is used for extracting an interaction fusion current waveform characteristic vector from the actual load current signal and the detection residual current signal; and
and the effective value generation module is used for generating the effective value of the residual current based on the interactive fusion current waveform characteristic vector.
According to another aspect of the present application, there is provided an acquisition method with residual current detection, including:
acquiring an actual load current signal and a detected residual current signal;
extracting an interactive fusion current waveform characteristic vector from the actual load current signal and the detected residual current signal; and
And generating an effective value of the residual current based on the interactive fusion current waveform characteristic vector.
Compared with the prior art, the acquisition terminal with the residual current detection, provided by the application, has the advantages that the intelligent calculation of the effective value of the residual current is realized on the basis of the actual load current and the detected residual current, the accuracy of the effective value of the residual current is improved, the residual current detection scheme is optimized, and meanwhile, the dependence on a manually specified correction parameter set is reduced.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
FIG. 1 is a block diagram of an acquisition terminal with residual current detection according to an embodiment of the present application;
FIG. 2 is a system architecture diagram of an acquisition terminal with residual current detection according to an embodiment of the present application;
FIG. 3 is a block diagram of a current signature interaction module in an acquisition terminal with residual current detection according to an embodiment of the present application;
FIG. 4 is a block diagram of an active value generation module in an acquisition terminal with residual current detection according to an embodiment of the present application;
FIG. 5 is a block diagram of an information gain unit in an acquisition terminal with residual current detection according to an embodiment of the present application;
FIG. 6 is a flow chart of an acquisition method with residual current detection according to an embodiment of the application;
fig. 7 is a schematic view of a scenario of an acquisition terminal with residual current detection according to an embodiment of the present application.
Detailed Description
Hereinafter, exemplary 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 some 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 by the example embodiments described herein.
As used in the specification and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.
A flowchart is used in the present application to describe the operations performed by a system according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.
Hereinafter, exemplary 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 some 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 by the example embodiments described herein.
In patent CN113721070a method of detecting the residual current value is provided, but the correction parameters in this method depend on a look-up table operation in a memory. The correction parameter set is selected based on manual regulation, and if the correction parameter set has an inaccurate or imperfect problem, the detection of the residual current has a great adverse effect. Therefore, an optimized residual current detection scheme is desired.
In the technical scheme of the application, an acquisition terminal with residual current detection is provided. Fig. 1 is a block diagram of an acquisition terminal with residual current detection according to an embodiment of the present application. Fig. 2 is a system architecture diagram of an acquisition terminal with residual current detection according to an embodiment of the present application. As shown in fig. 1 and 2, an acquisition terminal 300 with residual current detection according to an embodiment of the present application includes: a current signal acquisition module 310 for acquiring an actual load current signal and detecting a residual current signal; a current feature interaction module 320, configured to extract an interaction fusion current waveform feature vector from the actual load current signal and the detected residual current signal; and an effective value generating module 330, configured to generate an effective value of the residual current based on the feature vector of the inter-fusion current waveform.
Specifically, the current signal acquisition module 310 is configured to acquire an actual load current signal and detect a residual current signal. The load current signal refers to a current signal flowing through a load in a circuit. The load may be a resistor, inductor, capacitor, or other electronic component. Measuring load current signals is important for circuit performance analysis and fault diagnosis. By measuring the load current signal, the operating state of the load in the circuit, the power consumption, and possible faults or anomalies can be known. The residual current signal refers to the current in the circuit that does not pass through the primary load. In power systems, residual current signals are often used to detect faults or unbalance conditions in the circuit. Typically, the residual current signal is obtained by comparing the main load current with the total current.
According to an embodiment of the present application, an actual load current signal may be obtained by a current sensor; and acquiring a detection residual current signal through a current transformer. A current sensor is a device for measuring the current in an electrical circuit that converts the current into an electrical signal that can be measured. The function of the current sensor is to monitor the current change in the circuit and convert it into a signal suitable for use by the measurement and control system. Current sensors find wide application in many fields including power systems, industrial automation, electric vehicles, and electronic devices. They can be used for applications such as current measurement, current protection, electrical energy metering and current control. A current transformer is a device for measuring and monitoring the current in a circuit. It is a power transformer that converts high current (main load current) to low current for ease of measurement and use of the protection device. The primary function of the current transformer is to convert high current into low current for connection to a measuring instrument, protective equipment or control system for monitoring and control. It provides an isolated measurement loop so that the measurement device can safely acquire the current value in the circuit without being affected by high currents.
Specifically, the current feature interaction module 320 is configured to extract an interaction fusion current waveform feature vector from the actual load current signal and the detected residual current signal. In particular, in one specific example of the present application, as shown in fig. 3, the current feature interaction module 320 includes: the noise reduction unit 321 is configured to perform noise reduction processing on the actual load current signal and the detected residual current signal to obtain a noise-reduced actual load current signal and a noise-reduced detected residual current signal; the waveform feature extraction unit 322 is configured to extract waveform features of the waveform graph of the denoised actual load current signal and the waveform graph of the denoised detected residual current signal to obtain an actual load current waveform feature vector and a detected residual current waveform feature vector; and a feature interaction unit 323, configured to perform feature interaction on the actual load current waveform feature vector and the detected residual current waveform feature vector to obtain the interaction fusion current waveform feature vector.
More specifically, the noise reduction unit 321 is configured to perform noise reduction processing on the actual load current signal and the detected residual current signal to obtain a noise reduced actual load current signal and a noise reduced detected residual current signal. It is considered that in a practical scenario, the current signal often has various types of noise, which may come from fluctuations in the power supply, electromagnetic interference, noise of the sensor itself, etc. For example, high frequency noise due to electromagnetic radiation, switching operations, or other devices; as another example, fluctuations from the power supply or instability of the power system lead to slow drift of the current signal. Therefore, in the technical scheme of the application, noise interference can be removed by adopting noise reduction treatment so as to improve the accuracy and reliability of signals.
Accordingly, in one possible implementation, the noise reduction processing may be performed on the actual load current signal and the detected residual current signal to obtain a noise reduced actual load current signal and a noise reduced detected residual current signal, for example: collecting the original data of an actual load current signal and a detected residual current signal; the raw data is preprocessed, including the removal of noise and interference that may be present. May be implemented using filters, denoising algorithms, or other signal processing techniques; an appropriate noise reduction method is selected based on the characteristics of the actual load current signal and the detected residual current signal. Common noise reduction methods include mean filtering, median filtering, wavelet transformation, etc.; applying a selected noise reduction method to the actual load current signal and the detected residual current signal to reduce the effects of noise and interference; and checking whether the noise-reduced signal meets the requirement. The method can be used for evaluating the noise-reduced signal by comparing the statistical characteristics such as peak value, mean value, variance and the like of the noise-reduced signal with the original signal; if the noise reduction effect is not ideal, the noise reduction parameter can be tried to be adjusted or other noise reduction methods can be selected for optimization; and carrying out subsequent analysis and processing, such as power calculation, fault detection and the like, according to the actual load current signal after noise reduction and the detected residual current signal after noise reduction.
More specifically, the waveform feature extraction unit 322 is configured to perform waveform feature extraction on the waveform diagram of the actual load current signal after noise reduction and the waveform diagram of the detected residual current signal after noise reduction to obtain an actual load current waveform feature vector and a detected residual current waveform feature vector, respectively. That is, a waveform feature extractor is constructed using a convolutional neural network model to extract local neighborhood correlation pattern features for the current signal in the waveform of the actual load current signal after noise reduction and the waveform of the detected residual current signal after noise reduction. In particular, in one specific example of the present application, the waveform feature extraction unit 322 includes: the first convolution subunit is used for obtaining the waveform characteristic vector of the actual load current through a waveform characteristic extractor based on a first convolution neural network model according to the waveform graph of the actual load current signal after noise reduction; and the second convolution subunit is used for enabling the waveform graph of the residual current detection signal after noise reduction to pass through a waveform feature extractor based on a second convolution neural network model so as to obtain the residual current detection waveform feature vector.
The first convolution subunit is configured to obtain the waveform characteristic vector of the actual load current waveform by using a waveform characteristic extractor based on a first convolution neural network model according to the waveform graph of the actual load current signal after noise reduction. Specifically, each layer of the waveform characteristic extractor based on the first convolutional neural network model is used for respectively carrying out input data in forward transfer of the layer: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on a feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the waveform characteristic extractor based on the first convolutional neural network model is the waveform characteristic vector of the actual load current, and the input of the first layer of the waveform characteristic extractor based on the first convolutional neural network model is the waveform diagram of the actual load current signal after noise reduction.
Convolutional neural network (Convolutional Neural Network, CNN for short) is a deep learning model, mainly used for image recognition and computer vision tasks. The method has great success in the field of image processing, and is widely applied to tasks such as object detection, image classification, face recognition and the like. The core idea of CNN is to extract and learn the features of the image through the convolution layer, pooling layer and full-connection layer. The convolution layer convolves the input image with a set of learnable filters (also called convolution kernels) to extract local features in the image. The pooling layer is used for reducing the dimension of the feature map, reducing the calculation amount and extracting more robust features. Finally, the fully connected layer maps the extracted features onto output categories.
The second convolution subunit is configured to pass the waveform diagram of the noise-reduced detected residual current signal through a waveform feature extractor based on a second convolution neural network model to obtain the waveform feature vector of the detected residual current. Specifically, each layer of the waveform feature extractor based on the second convolutional neural network model is used for respectively carrying out forward transfer on input data: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on a feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the waveform characteristic extractor based on the second convolutional neural network model is the waveform characteristic vector of the detected residual current, and the input of the first layer of the waveform characteristic extractor based on the second convolutional neural network model is the waveform diagram of the detected residual current signal after noise reduction.
It should be noted that, in other specific examples of the present application, the waveform feature extraction may be further performed on the waveform diagram of the actual load current signal after noise reduction and the waveform diagram of the detected residual current signal after noise reduction in other manners to obtain an actual load current waveform feature vector and a detected residual current waveform feature vector, for example: extracting waveform characteristics of actual load current signals: collecting an actual load current signal; noise reduction processing is performed on the signal, for example, a filter or a denoising algorithm is used to remove noise and interference; waveform characteristics of the noise-reduced signal are extracted, and the following characteristics can be considered: peak value: maximum or minimum, average of recorded signal: calculating the mean value and variance of the signals: the degree of dispersion of the signal, root mean square value (RMS): calculating root mean square value of the signal, representing effective value and waveform factor of the signal: the peak to RMS ratio of the signal is calculated to describe the ripple, harmonic content of the signal: analyzing harmonic components in the signal, the amplitude and phase of the harmonic can be calculated, and the frequency spectrum analysis is carried out: converting the signal into a frequency domain, and analyzing energy distribution of different frequency components; extracting waveform characteristics of the detected residual current signal: collecting and detecting residual current signals; noise reduction processing is carried out on the signals so as to remove noise and interference; waveform characteristics of the noise-reduced signal are extracted, and the following characteristics can be considered: peak value: maximum or minimum, average of recorded signal: calculating the mean value and variance of the signals: the degree of dispersion of the signal, root mean square value (RMS): the root mean square value of the signal is calculated, representing the effective value of the signal. Waveform factor: the peak to RMS ratio of the signal is calculated to describe the ripple, harmonic content of the signal: analyzing harmonic components in the signal, the amplitude and phase of the harmonic can be calculated, and the frequency spectrum analysis is carried out: the signal is converted to the frequency domain and the energy distribution of the different frequency components is analyzed.
More specifically, the feature interaction unit 323 is configured to perform feature interaction on the actual load current waveform feature vector and the detected residual current waveform feature vector to obtain the interaction fusion current waveform feature vector. As previously mentioned, the effective value of the residual current depends on the difference between the actual load current and the residual current. In other words, the interactions and associations between the actual load current and the residual current affect the accuracy of the decoding of the effective value of the residual current by the subsequent model. Therefore, in the technical scheme of the application,further, a cascading function is used to interpolate the actual load current waveform feature vector and the detected residual current waveform feature vector to obtain an interpolated fused current waveform feature vector. The cascading function can enable the network to have certain logic reasoning capacity through point convolution and activation operation, and related information and interaction relation among vectors are mined, so that the interaction fusion current waveform characteristic vector can more accurately represent interaction effect and interaction characteristics between the actual load current waveform characteristic vector and the detection residual current waveform characteristic vector. Specifically, in one specific example of the present application, the actual load current waveform feature vector and the detected residual current waveform feature vector are interacted using a cascading function to obtain the interacted fusion current waveform feature vector; wherein, the formula is: f (X) i ,X j )=Relu(W f [θ(X i ),φ(X j )]) Wherein W is f ,θ(X i ) And phi (X) j ) All representing the point convolution of the input as an activation function, []Representing the splicing operation, X i X is the characteristic value of each position in the characteristic vector of the actual load current waveform j And the characteristic value of each position in the characteristic vector of the detected residual current waveform is obtained.
It should be noted that, in other specific examples of the present application, the actual load current waveform feature vector and the detected residual current waveform feature vector may be further subjected to feature interaction in other manners to obtain the interaction fusion current waveform feature vector, for example: collecting actual load current waveform characteristic vectors: first, a current waveform signal of an actual load is acquired by a current sensor. This current waveform signal may be a continuous analog signal or a discrete digital signal. The current waveform signal is then sampled and a set of eigenvectors is calculated. These feature vectors may include averages, peaks, spectral distributions, etc.; collecting and detecting residual current waveform characteristic vectors: next, a waveform signal for detecting the residual current is acquired by a current sensor. Again, this current waveform signal may be a continuous analog signal or a discrete digital signal. The current waveform signal is sampled and a set of eigenvectors is calculated. These feature vectors may include averages, peaks, spectral distributions, etc.; feature interaction: and carrying out characteristic interaction on the actual load current waveform characteristic vector and the detected residual current waveform characteristic vector. This may be achieved by various methods, such as stitching, weighted summing, differencing, etc. the two sets of feature vectors. The method aims at fusing the two groups of characteristic vectors together to obtain a new interactive fusion current waveform characteristic vector; obtaining an interactive fusion current waveform characteristic vector: and processing the actual load current waveform characteristic vector and the detected residual current waveform characteristic vector according to the selected characteristic interaction method to obtain an interaction fusion current waveform characteristic vector. This eigenvector will contain information from both sets of eigenvectors reflecting the relationship between the actual load current and the detected residual current.
It should be noted that, in other specific examples of the present application, the feature vector of the inter-fusion current waveform may be extracted from the actual load current signal and the detected residual current signal in other manners, for example: acquiring an actual load current signal: the actual load current signal is obtained from the power system or other current source. This may be achieved by connecting the current sensor to the load circuit; acquiring a detection residual current signal: a current sensor or other detection device is used to obtain the residual current signal. Residual current refers to current that is not absorbed by the actual load, which may be present due to current leakage or other reasons; and (3) interaction fusion: and carrying out interactive fusion on the actual load current signal and the detected residual current signal. This may be accomplished by superimposing, subtracting, or otherwise processing the two signals; extracting current waveform characteristics: and extracting the characteristics of the current waveform from the signals after the mutual fusion. This may include amplitude, frequency, phase difference, harmonic content, etc.; constructing a feature vector: and combining the extracted current waveform characteristics into a characteristic vector. The feature vector is a vector containing a plurality of feature values for describing the features of the current waveform.
Specifically, the effective value generating module 330 is configured to generate an effective value of the residual current based on the feature vector of the inter-fusion current waveform. In particular, in one specific example of the present application, as shown in fig. 4, the valid value generation module 330 includes: the information gain unit 331 is configured to perform information gain on the inter-fusion current waveform feature vector to obtain an optimized inter-fusion current waveform feature vector; and a decoding regression unit 332, configured to perform decoding regression on the optimized inter-fusion current waveform feature vector through a decoder to obtain a decoded value, where the decoded value is used to represent an effective value of the residual current.
More specifically, the information gain unit 331 is configured to perform information gain on the inter-integrated current waveform feature vector to obtain an optimized inter-integrated current waveform feature vector. In particular, in one specific example of the present application, as shown in fig. 5, the information gain unit 331 includes: an information preserving and fusing subunit 3311, configured to perform forward propagation information preserving and fusing on the actual load current waveform feature vector and the detected residual current waveform feature vector to obtain a corrected feature vector; and an information gain subunit 3312, configured to gain information on the inter-fusion current waveform feature vector based on the correction feature vector to obtain the optimized inter-fusion current waveform feature vector.
The information preserving and fusing subunit 3311 is configured to perform forward propagation information preserving and fusing on the actual load current waveform feature vector and the detected residual current waveform feature vector to obtain a corrected feature vector. In the technical scheme of the application, the actual load current waveform feature vector and the detected residual current waveform feature vector respectively express the image semantic features of the waveform diagrams of the actual load current signal after noise reduction and the detected residual current signal after noise reduction, so that the image semantic feature distribution of the actual load current waveform feature vector and the detected residual current waveform feature vector cannot be completely misaligned due to the difference of source image semantics. Thus, the inter-fusion current waveform characteristic vector is obtained by fusing the actual load current waveform characteristic vector and the detected residual current waveform characteristic vector by using a cascading functionWhen the feature vector is represented, the respective misaligned image semantic feature distributions of the actual load current waveform feature vector and the detected residual current waveform feature vector generate information loss during forward propagation of the model when point convolution and activation operations are performed through cascading functions, and the accuracy of a decoding result obtained by the interactive fusion current waveform feature vector through a decoder is affected. Based on this, the applicant of the present application has identified the actual load current waveform feature vector, e.g., as V 1 And said detected residual current waveform feature vector, e.g. denoted as V 2 Forward propagation information preserving fusion is performed to obtain a corrected feature vector V ', where V' is expressed as:
"s" and "s" denote shifting the feature vector left by s bits and right by s bits, respectively, round is a rounding function,is the characteristic vector V of the actual load current waveform 1 And the detected residual current waveform characteristic vector V 2 Is the average of all feature values of (i) i· (i) 1 Represents a norm, d (V) 1 ,V 2 ) Is the characteristic vector V of the actual load current waveform 1 And the detected residual current waveform characteristic vector V 2 The distance between them, and log is the base 2 logarithm. Here, the characteristic vector V for the actual load current waveform 1 And the detected residual current waveform characteristic vector V 2 In the forward propagation process in the network model, floating point distribution errors and information loss on vector scale due to convolution and activation operations balance and normalize the quantities in the forward propagation process by introducing a bitwise displacement operation of the vector from the point of view of unification informationError and information loss are converted, and distribution diversity is introduced by reshaping the distribution of characteristic parameters before fusion, thereby information retention (extension) is performed in a manner that enlarges the information entropy. Therefore, after the correction feature vector V' is subjected to linear interpolation to be converted into the same length as the cross fusion current waveform feature vector, the cross fusion current waveform feature vector is subjected to dot-multiplication weighting, so that the information loss of the cross fusion current waveform feature vector can be reduced, and the accuracy of a decoding result obtained by the cross fusion current waveform feature vector through a decoder is improved.
The information gain subunit 3312 is configured to perform information gain on the inter-fusion current waveform feature vector based on the correction feature vector to obtain the optimized inter-fusion current waveform feature vector. In particular, the information gain subunit 3312 includes: a linear interpolation secondary subunit, configured to obtain a corrected feature vector after length adjustment by performing linear interpolation on the corrected feature vector, where the corrected feature vector after length adjustment and the interactive fusion current waveform feature vector have the same length; and the weighting secondary subunit is used for carrying out dot multiplication weighting on the interaction fusion current waveform characteristic vector based on the length-adjusted correction characteristic vector so as to obtain the optimized interaction fusion current waveform characteristic vector.
The linear interpolation secondary subunit is configured to obtain a corrected feature vector after length adjustment by performing linear interpolation on the corrected feature vector, where the corrected feature vector after length adjustment and the interactive fusion current waveform feature vector have the same length. It is noted that linear interpolation is a common interpolation method for estimating the values of other data points between given data points. It is based on a simple assumption that the data between two known data points varies linearly. Linear interpolation is suitable for situations where the trend of change between data points is relatively smooth and the rate of change is slow. It is widely used in applications such as image processing, data analysis, numerical simulation, and the like.
Accordingly, in one possible implementation, the corrected feature vector after length adjustment may be obtained by linearly interpolating the corrected feature vector, where the corrected feature vector after length adjustment and the inter-fused current waveform feature vector have the same length, for example: acquiring an original correction characteristic vector and an interactive fusion current waveform characteristic vector; determining the lengths of the correction characteristic vector and the interactive fusion current waveform characteristic vector; if the lengths of the two feature vectors are the same, interpolation adjustment is not required, and they already have the same length; if the lengths of the two feature vectors are different, interpolation adjustment is needed; and determining the interpolation step length according to the length difference required to be adjusted. The step length can be adjusted according to the needs, and common interpolation methods comprise linear interpolation, polynomial interpolation and the like; performing interpolation adjustment on the correction characteristic vector by using a linear interpolation method to enable the correction characteristic vector to have the same length as the interactive fusion current waveform characteristic vector; after interpolation adjustment is completed, a correction feature vector with the length adjusted is obtained; and finally, performing subsequent processing and analysis by using the corrected characteristic vector and the interactive fusion current waveform characteristic vector after the length adjustment.
And the weighting secondary subunit is used for carrying out dot-multiplication weighting on the interaction fusion current waveform characteristic vector based on the length-adjusted correction characteristic vector so as to obtain the optimized interaction fusion current waveform characteristic vector. Accordingly, in one possible implementation, the cross-fused current waveform feature vector may be obtained by performing point multiplication weighting on the cross-fused current waveform feature vector based on the length-adjusted correction feature vector by: and (3) length adjustment: first, the correction feature vector is length-adjusted. This may involve interpolating or truncating the vector to match it to the length of the inter-fused current waveform feature vector; dot product of correction feature vector and interaction fusion current waveform feature vector: and performing dot multiplication operation on the correction characteristic vector and the interactive fusion current waveform characteristic vector. Dot multiplication refers to multiplying elements at positions corresponding to two vectors and adding the results; weighting: and weighting the dot multiplication result. This involves assigning a weight to each element in the dot product. The weight can be selected according to specific requirements and algorithm design; optimizing the feature vector of the cross fusion current waveform: and adding the weighted result with the feature vector of the cross fusion current waveform to obtain the optimized feature vector of the cross fusion current waveform.
It should be noted that, in other specific examples of the present application, the information gain may be performed on the inter-fusion current waveform feature vector based on the correction feature vector in other manners to obtain the optimized inter-fusion current waveform feature vector, for example: and obtaining a correction feature vector. The correction feature vector is obtained by measuring and analyzing a known current waveform. These known current waveforms may be measured under standard conditions to provide accurate reference values; and obtaining an inter-fusion current waveform characteristic vector, wherein the inter-fusion current waveform characteristic vector is obtained by fusing a plurality of current waveforms. These current waveforms may be obtained from different current sensors or from different locations or time periods of the same current sensor; information gain is calculated, and the information gain is an index for measuring the contribution degree of the characteristics to classification or regression tasks. In this case, the information gain may be used to evaluate the extent to which the correction feature vector contributes to optimizing the cross-fused current waveform feature vector; the interactive fusion current waveform feature vector is optimized, and the optimization process can select to reserve or discard specific features according to the magnitude of the information gain so as to improve the quality and efficiency of the final feature vector.
It should be noted that, in other specific examples of the present application, the information gain may be performed on the inter-fusion current waveform feature vector in other manners to obtain an optimized inter-fusion current waveform feature vector, for example: collecting original interaction fusion current waveform data: first, the required inter-fusion current waveform data needs to be collected. This can be done by acquiring a current signal using a current sensor and recording it; extracting a current waveform characteristic vector: extracting feature vectors from raw current waveform data is a key step in optimization. Various feature extraction methods may be used, such as time domain features, frequency domain features, or wavelet transforms. These features may include peaks, means, variances, spectral distributions, etc.; calculating information gain: information gain is a measure of the importance of an evaluation feature to a classification or prediction task. In this case, we want to determine which features are most important for optimizing the inter-fused current waveform feature vector by calculating the information gain. The information gain may be calculated using different algorithms, such as decision tree algorithms or entropy calculation methods; selecting an optimized feature vector: and selecting the most important characteristic for optimizing the feature vector of the interactive fusion current waveform according to the information gain obtained by calculation. These features will constitute the final optimized feature vector; applying the optimized feature vector: the optimized feature vector is applied to the relevant task or application. This may involve sorting, prediction, monitoring, etc. operations, depending on the particular field of application.
More specifically, the decoding regression unit 332 is configured to perform decoding regression on the optimized inter-fusion current waveform feature vector by using a decoder to obtain a decoded value, where the decoded value is used to represent an effective value of the residual current. That is, a decoder is used to perform a decoding regression operation on the inter-fusion current waveform feature vector, thereby achieving an automated calculation of the effective value of the residual current. More specifically, the optimized inter-fusion current waveform feature vector is subjected to a decoding regression with the following formula using the decoder to obtain a decoded value representing an effective value of the residual current; wherein, the formula is:
wherein X represents the optimized cross fusion current waveform characteristic vector, Y is the decoding value, W is a weight matrix,>representing matrix multiplication.
Decoding regression refers to the process of re-restoring already encoded data to the original data. In machine learning, a regression problem refers to predicting the value of a continuous output variable from a given input variable. Decoding regression is the re-reduction of encoded input data to original input data for further analysis or application. Decoding regression has a wide range of applications in practical applications. For example, in image processing, the encoded image data may be restored to the original image using a decoding regression. In text processing, the encoded text data may be restored to the original text content using a decoding regression. Decoding regression can also be used in the fields of data compression, data recovery, and the like.
It should be noted that, in other specific examples of the present application, the effective value of the residual current may also be generated by other manners based on the feature vector of the inter-fusion current waveform, for example: collecting current waveform data: first, current waveform data to be processed needs to be collected. The current signal may be acquired by a sensor or a measuring device; feature extraction: features are extracted from the collected current waveform data. Common characteristics include amplitude, frequency, phase, etc. These features may be extracted by mathematical algorithms or signal processing techniques; feature vector generation: the extracted features are combined into feature vectors. The feature vector is a vector containing a plurality of feature values for representing the features of the current waveform; and (3) interaction fusion: and carrying out interactive fusion on the plurality of feature vectors. This may be achieved by statistical methods, neural networks or other machine learning algorithms. The purpose of interactive fusion is to comprehensively utilize the information of different feature vectors, and improve the accuracy and stability of the effective value of the residual current; calculating the effective value of the residual current: and calculating the effective value of the residual current by using a proper algorithm according to the feature vector after the interaction fusion. Common calculation methods include Root Mean Square (RMS) method, integral method, and the like; and (3) outputting results: and outputting the calculated residual current effective value. The results may be displayed on a screen, saved to a file or transmitted to other devices for further processing.
As described above, the acquisition terminal 300 with residual current detection according to the embodiment of the present application may be implemented in various wireless terminals, such as a server or the like having an acquisition algorithm with residual current detection. In one possible implementation, the acquisition terminal 300 with residual current detection according to embodiments of the present application may be integrated into a wireless terminal as a software module and/or hardware module. For example, the acquisition terminal 300 with residual current detection may be a software module in the operating system of the wireless terminal, or may be an application developed for the wireless terminal; of course, the acquisition terminal 300 with residual current detection may also be one of a number of hardware modules of the wireless terminal.
Alternatively, in another example, the acquisition terminal 300 with residual current detection and the wireless terminal may be separate devices, and the acquisition terminal 300 with residual current detection may be connected to the wireless terminal through a wired and/or wireless network and transmit the interaction information in a agreed data format.
Further, an acquisition method with residual current detection is also provided.
Fig. 6 is a flowchart of an acquisition method with residual current detection according to an embodiment of the present application. . As shown in fig. 6, an acquisition method with residual current detection according to an embodiment of the present application includes: s110, acquiring an actual load current signal and a detected residual current signal; s120, extracting an interactive fusion current waveform characteristic vector from the actual load current signal and the detected residual current signal; and S130, generating an effective value of the residual current based on the cross fusion current waveform characteristic vector.
In summary, the acquisition method with residual current detection according to the embodiment of the application is explained, which achieves intelligent calculation of the effective value of the residual current by based on the actual load current and the detected residual current, improves the accuracy of the effective value of the residual current, optimizes the residual current detection scheme, and simultaneously reduces the dependence on the manually specified correction parameter set.
Fig. 7 is a schematic view of a scenario of an acquisition terminal with residual current detection according to an embodiment of the present application. . As shown in fig. 7, in this application scenario, the actual load current signal is acquired by a current sensor (e.g., V1 as illustrated in fig. 7); the detected residual current signal is acquired by a current transformer (e.g., V2 as illustrated in fig. 7). The signal is then input to a server (e.g., S in fig. 7) that is deployed with an acquisition algorithm for residual current detection, wherein the server is capable of processing the input signal with the acquisition algorithm for residual current detection to generate a decoded value that is indicative of the effective value of residual current.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. The utility model provides a take residual current detection's acquisition terminal which characterized in that includes:
the current signal acquisition module is used for acquiring an actual load current signal and detecting a residual current signal;
the current characteristic interaction module is used for extracting an interaction fusion current waveform characteristic vector from the actual load current signal and the detection residual current signal; and
and the effective value generation module is used for generating the effective value of the residual current based on the interactive fusion current waveform characteristic vector.
2. The acquisition terminal with residual current detection according to claim 1, wherein the current feature interaction module comprises:
The noise reduction unit is used for carrying out noise reduction processing on the actual load current signal and the detected residual current signal to obtain a noise-reduced actual load current signal and a noise-reduced detected residual current signal;
the waveform characteristic extraction unit is used for extracting waveform characteristics of the waveform diagram of the actual load current signal after noise reduction and the waveform diagram of the residual current signal after noise reduction respectively to obtain an actual load current waveform characteristic vector and a residual current waveform characteristic vector; and
and the characteristic interaction unit is used for carrying out characteristic interaction on the actual load current waveform characteristic vector and the detected residual current waveform characteristic vector so as to obtain the interaction fusion current waveform characteristic vector.
3. The acquisition terminal with residual current detection according to claim 2, wherein the waveform feature extraction unit is configured to:
the first convolution subunit is used for obtaining the waveform characteristic vector of the actual load current through a waveform characteristic extractor based on a first convolution neural network model according to the waveform graph of the actual load current signal after noise reduction; and
and the second convolution subunit is used for obtaining the waveform characteristic vector of the detected residual current by passing the waveform diagram of the noise-reduced detected residual current signal through a waveform characteristic extractor based on a second convolution neural network model.
4. The acquisition terminal with residual current detection according to claim 3, wherein the feature interaction unit is configured to:
using a cascading function to interact the actual load current waveform feature vector and the detected residual current waveform feature vector to obtain the interaction fusion current waveform feature vector;
wherein, the formula is:
f(X i ,X j )=Relu(W f [θ(X i ),φ(X j )])
wherein W is f ,θ(X i ) And phi (X) j ) All representing the point convolution of the input as an activation function, []Representing the splicing operation, X i X is the characteristic value of each position in the characteristic vector of the actual load current waveform j And the characteristic value of each position in the characteristic vector of the detected residual current waveform is obtained.
5. The acquisition terminal with residual current detection according to claim 4, wherein the effective value generation module comprises:
the information gain unit is used for carrying out information gain on the interactive fusion current waveform characteristic vector so as to obtain an optimized interactive fusion current waveform characteristic vector; and
and the decoding regression unit is used for carrying out decoding regression on the optimized interactive fusion current waveform characteristic vector through a decoder to obtain a decoding value, wherein the decoding value is used for representing the effective value of the residual current.
6. The acquisition terminal with residual current detection according to claim 5, wherein the information gain unit comprises:
the information retention and fusion subunit is used for carrying out forward propagation information retention and fusion on the actual load current waveform characteristic vector and the detected residual current waveform characteristic vector so as to obtain a correction characteristic vector; and
and the information gain subunit is used for carrying out information gain on the interaction fusion current waveform characteristic vector based on the correction characteristic vector so as to obtain the optimized interaction fusion current waveform characteristic vector.
7. The acquisition terminal with residual current detection according to claim 6, wherein the information retention fusion subunit is configured to: carrying out forward propagation information retention fusion on the actual load current waveform characteristic vector and the detected residual current waveform characteristic vector by using the following formula to obtain a correction characteristic vector;
wherein, the formula is:
"s" and "s" denote shifting the feature vector left by s bits and right by s bits, respectively, round is a rounding function,is the average value of all characteristic values of the actual load current waveform characteristic vector and the detected residual current waveform characteristic vector 1 Represents a norm, d (V) 1 ,V 2 ) Is the distance between the actual load current waveform feature vector and the detected residual current waveform feature vector, and log is the base 2 logarithm.
8. The acquisition terminal with residual current detection according to claim 7, wherein the information gain subunit comprises:
a linear interpolation secondary subunit, configured to obtain a corrected feature vector after length adjustment by performing linear interpolation on the corrected feature vector, where the corrected feature vector after length adjustment and the interactive fusion current waveform feature vector have the same length; and
and the weighted secondary subunit is used for carrying out point multiplication weighting on the interaction fusion current waveform characteristic vector based on the corrected characteristic vector after the length adjustment so as to obtain the optimized interaction fusion current waveform characteristic vector.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310864568.7A CN116908518B (en) | 2023-07-13 | 2023-07-13 | Acquisition terminal with residual current detection function |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310864568.7A CN116908518B (en) | 2023-07-13 | 2023-07-13 | Acquisition terminal with residual current detection function |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116908518A true CN116908518A (en) | 2023-10-20 |
CN116908518B CN116908518B (en) | 2024-04-05 |
Family
ID=88354322
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310864568.7A Active CN116908518B (en) | 2023-07-13 | 2023-07-13 | Acquisition terminal with residual current detection function |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116908518B (en) |
Citations (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1309772A (en) * | 1998-06-30 | 2001-08-22 | 德耳塔电气有限公司 | Residual current detection device |
US20050212505A1 (en) * | 2001-06-08 | 2005-09-29 | Murray Martin A | Measuring devices |
CN102135555A (en) * | 2010-12-29 | 2011-07-27 | 重庆大学 | Series arcing fault identifying method for low-voltage system |
EP2846432A1 (en) * | 2013-09-05 | 2015-03-11 | Siemens Aktiengesellschaft | Residual current detection method and device |
CN105281291A (en) * | 2015-11-23 | 2016-01-27 | 上海电机学院 | Residual current protective device and current protective method thereof |
CN107578016A (en) * | 2017-09-06 | 2018-01-12 | 重庆大学 | A kind of residual current waveform automatic identifying method based on rarefaction representation |
US20180196091A1 (en) * | 2017-01-06 | 2018-07-12 | Liebert Corporation | System and method of identifying path of residual current flow through an intelligent power strip |
US20180299499A1 (en) * | 2015-10-07 | 2018-10-18 | Jenoptik Advanced Systems Gmbh | Fault current protection device for monitoring an electric load for a vehicle, and method for carrying out a self-test of a fault current sensor |
CN110601582A (en) * | 2018-06-13 | 2019-12-20 | 通用电气能源能量变换技术有限公司 | Optimal output control of alternating current in a multi-power stacked inverter |
CN111122956A (en) * | 2019-09-04 | 2020-05-08 | 天津市中力神盾电子科技有限公司 | Residual current detection and correction device, residual current detection and correction method and electrical fire monitoring system |
CN212540499U (en) * | 2019-12-02 | 2021-02-12 | 天津市中力神盾电子科技有限公司 | Device for detecting residual current detector |
CN213302351U (en) * | 2020-10-30 | 2021-05-28 | 国网重庆市电力公司检修分公司 | Residual current collecting device of low-voltage alternating-current power supply system of transformer substation |
CN113552406A (en) * | 2021-07-27 | 2021-10-26 | 上海电机学院 | High-precision residual current detection device powered by single power supply |
CN113721070A (en) * | 2020-05-26 | 2021-11-30 | 天津首瑞智能电气有限公司 | Residual current detection device and residual current detection method |
DE102022129480A1 (en) * | 2021-11-18 | 2023-05-25 | Elmos Semiconductor Se | Electronic security with encrypted and compressed data communication and its application and further training |
US20230184812A1 (en) * | 2021-12-10 | 2023-06-15 | Vertiv Corporation | Residual current monitoring type b with integrated self-test system and method |
-
2023
- 2023-07-13 CN CN202310864568.7A patent/CN116908518B/en active Active
Patent Citations (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1309772A (en) * | 1998-06-30 | 2001-08-22 | 德耳塔电气有限公司 | Residual current detection device |
US20050212505A1 (en) * | 2001-06-08 | 2005-09-29 | Murray Martin A | Measuring devices |
CN102135555A (en) * | 2010-12-29 | 2011-07-27 | 重庆大学 | Series arcing fault identifying method for low-voltage system |
EP2846432A1 (en) * | 2013-09-05 | 2015-03-11 | Siemens Aktiengesellschaft | Residual current detection method and device |
US20180299499A1 (en) * | 2015-10-07 | 2018-10-18 | Jenoptik Advanced Systems Gmbh | Fault current protection device for monitoring an electric load for a vehicle, and method for carrying out a self-test of a fault current sensor |
CN105281291A (en) * | 2015-11-23 | 2016-01-27 | 上海电机学院 | Residual current protective device and current protective method thereof |
CN110168391A (en) * | 2017-01-06 | 2019-08-23 | 维谛公司 | The system and method for being identified by the path of the residual current of intelligent power plate |
US20180196091A1 (en) * | 2017-01-06 | 2018-07-12 | Liebert Corporation | System and method of identifying path of residual current flow through an intelligent power strip |
CN107578016A (en) * | 2017-09-06 | 2018-01-12 | 重庆大学 | A kind of residual current waveform automatic identifying method based on rarefaction representation |
CN110601582A (en) * | 2018-06-13 | 2019-12-20 | 通用电气能源能量变换技术有限公司 | Optimal output control of alternating current in a multi-power stacked inverter |
CN111122956A (en) * | 2019-09-04 | 2020-05-08 | 天津市中力神盾电子科技有限公司 | Residual current detection and correction device, residual current detection and correction method and electrical fire monitoring system |
CN212540499U (en) * | 2019-12-02 | 2021-02-12 | 天津市中力神盾电子科技有限公司 | Device for detecting residual current detector |
CN113721070A (en) * | 2020-05-26 | 2021-11-30 | 天津首瑞智能电气有限公司 | Residual current detection device and residual current detection method |
CN213302351U (en) * | 2020-10-30 | 2021-05-28 | 国网重庆市电力公司检修分公司 | Residual current collecting device of low-voltage alternating-current power supply system of transformer substation |
CN113552406A (en) * | 2021-07-27 | 2021-10-26 | 上海电机学院 | High-precision residual current detection device powered by single power supply |
DE102022129480A1 (en) * | 2021-11-18 | 2023-05-25 | Elmos Semiconductor Se | Electronic security with encrypted and compressed data communication and its application and further training |
US20230184812A1 (en) * | 2021-12-10 | 2023-06-15 | Vertiv Corporation | Residual current monitoring type b with integrated self-test system and method |
Non-Patent Citations (1)
Title |
---|
张冠英;杨晓光;王尧;张波;: "一种智能剩余电流检测装置的开发", 低压电器, no. 02 * |
Also Published As
Publication number | Publication date |
---|---|
CN116908518B (en) | 2024-04-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6888564B2 (en) | Method and system for estimating sharpness metrics based on local edge kurtosis | |
Zygarlicki et al. | A reduced Prony's method in power-quality analysis—parameters selection | |
US7127364B2 (en) | Method of compensating for distorted secondary current of current transformer | |
Mehra et al. | Analysis of PCA based compression and denoising of smart grid data under normal and fault conditions | |
CN114879085B (en) | Single-phase earth fault identification method and device, electronic equipment and medium | |
CN111783696A (en) | Edge calculation method for low-voltage branch topology real-time analysis based on PV relation | |
CN117233682B (en) | Quick calibration system of balance bridge | |
CN117290788B (en) | Power distribution network fault identification method and system based on improved wavelet transformation algorithm | |
CN116908518B (en) | Acquisition terminal with residual current detection function | |
CN114169245A (en) | Transformer fault diagnosis method, device and equipment | |
CN117269644A (en) | Line fault monitoring system and method for current transformer | |
CN112418324A (en) | Cross-modal data fusion method for electrical equipment state perception | |
CN111353526A (en) | Image matching method and device and related equipment | |
CN114325072B (en) | Ferromagnetic resonance overvoltage identification method and device based on gram angular field coding | |
CN110794210B (en) | Method and device for judging voltage harmonic isolation effect, power supply system, computer equipment and storage medium | |
CN112422212B (en) | Data self-adaptive wireless communication channel prediction method, storage medium and equipment | |
CN114157023A (en) | Distribution transformer early warning information acquisition method | |
CN112485616A (en) | Cable insulation aging detection method and device, storage medium and processor | |
JP3620930B2 (en) | Power system characteristic estimation apparatus and characteristic estimation method | |
Medeiros et al. | TinyML Custom AI Algorithms for Low-Power IoT Data Compression: A Bridge Monitoring Case Study | |
CN112114215A (en) | Transformer aging evaluation method and system based on error back propagation algorithm | |
CN115542230B (en) | Current transformer error estimation method and device based on diffusion model | |
Jenkin | Fast Prediction of Contrast Detection Probability | |
CN115423221B (en) | Facility operation trend prediction method | |
CN115267462B (en) | Partial discharge type identification method based on self-adaptive label generation |
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 |