CN117713505B - High-side current detection method and system for switching power supply - Google Patents
High-side current detection method and system for switching power supply Download PDFInfo
- Publication number
- CN117713505B CN117713505B CN202410166774.5A CN202410166774A CN117713505B CN 117713505 B CN117713505 B CN 117713505B CN 202410166774 A CN202410166774 A CN 202410166774A CN 117713505 B CN117713505 B CN 117713505B
- Authority
- CN
- China
- Prior art keywords
- subsequence
- side current
- intrinsic
- eigenmode
- tuple
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 38
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 238000000354 decomposition reaction Methods 0.000 claims description 16
- 230000002123 temporal effect Effects 0.000 claims description 16
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000005355 Hall effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005672 electromagnetic field Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003071 parasitic effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B70/00—Technologies for an efficient end-user side electric power management and consumption
- Y02B70/10—Technologies improving the efficiency by using switched-mode power supplies [SMPS], i.e. efficient power electronics conversion e.g. power factor correction or reduction of losses in power supplies or efficient standby modes
Landscapes
- Measurement Of Current Or Voltage (AREA)
Abstract
The invention relates to the technical field of electric signal processing, in particular to a high-side current detection method and a system of a switching power supply, wherein the method comprises the following steps: collecting high-side current signals of a switching power supply, and decomposing to obtain each intrinsic mode; constructing extrinsic difference amplitude-frequency indexes of each intrinsic mode subsequence according to the zero number of each intrinsic mode subsequence and the upper envelope curve and the lower envelope curve; constructing a time intrinsic similarity index and a time intrinsic tuple heterogeneity of the subsequence; calculating the high-side current potential noise figure of each subsequence based on the calculated high-side current potential noise figure; and acquiring the window size of each subsequence when the moving average denoising algorithm denoises the subsequences according to the high-side current potential noise figure of each subsequence, and detecting the denoised high-side current sequence by combining a local outlier factor detection algorithm to finish the detection of the high-side current of the switching power supply. The invention carries out self-adaptive adjustment on the denoising window based on the potential noise figure of the high-side current, thereby realizing accurate detection on the high-side current of the switching power supply.
Description
Technical Field
The invention relates to the technical field of electric signal processing, in particular to a high-side current detection method and system for a switching power supply.
Background
A switching power supply is a common power conversion device that implements regulation of an output voltage by high frequency switching operation of switching devices and a transformer. In practical applications, in order to ensure the safety and stability of the switching power supply, the high-side current of the switching power supply needs to be detected and controlled. When the output current exceeds the normal set range, protective measures can be taken in time; the working frequency of the switch can be adjusted according to the current change so as to achieve the optimal working efficiency of the power supply; the abnormal current fluctuation can also be used as a reference basis for fault diagnosis.
In general, when the high-side detection is performed on the switching power supply, a current sensor is inserted into the high-side of the switching device, and the output current of the switching power supply is detected in real time. There are two common methods for high-side current detection: one is to adopt a resistor voltage division method, indirectly calculate the current by connecting a current sensor resistor at the high side of the switching device and measuring the voltage at the two ends of the current sensor resistor; the other is to directly measure the magnetic field change at the high side of the switching power supply by adopting a non-contact sensor such as a Hall effect or a magneto-resistance effect, so as to deduce the current change. However, the above method can generate some interference items when detecting the current, such as internal resistance, parasitic capacitance and inductance of the switch element, interference caused by external magnetic field change if a non-contact sensor is adopted, and large current change and high-frequency noise in the switching power supply system, which can generate electromagnetic interference to the measurement circuit, and an interference signal can be superimposed on the current signal to be measured.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a high-side current detection method and a system for a switching power supply, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for detecting a high-side current of a switching power supply, including the steps of:
collecting high-side current signals of a switching power supply to form an original high-side current sequence;
obtaining each intrinsic mode according to the original high-side current sequence; obtaining upper and lower envelope curves of each eigenmode according to the maximum and the minimum values of each eigenmode; dividing each eigenmode according to a time sequence to obtain each subsequence of each eigenmode, and constructing an extrinsic difference amplitude-frequency index of each eigenmode subsequence according to the number of zeros of each eigenmode subsequence and upper and lower envelope curves; the eigenvalue frequency index of the difference of the extrinsic of each subsequence in all eigenvectors is formed into the moment eigenvector of each subsequence; constructing a time intrinsic similarity index of the subsequences according to the time intrinsic tuple difference of the adjacent subsequences; constructing the time intrinsic tuple heterogeneity of each subsequence according to the element variance in the time intrinsic tuple of each subsequence and the similarity index between the time intrinsic tuples; obtaining a high-side current potential noise index of the subsequence according to the subsequence and the time intrinsic tuple heterogeneity of each subsequence in a preset neighborhood window;
and acquiring the window size of each subsequence when the moving average denoising algorithm denoises the subsequences according to the high-side current potential noise figure of each subsequence, denoising and reconstructing each subsequence to obtain a denoised high-side current sequence, and detecting the denoised high-side current sequence by combining a local outlier factor detection algorithm to finish the detection of the high-side current of the switching power supply.
Further, the obtaining each eigenmode according to the original high-side current sequence includes: and decomposing the original high-side current sequence by adopting an empirical mode decomposition algorithm, and outputting each intrinsic mode of the original high-side current sequence.
Further, the obtaining the upper envelope line and the lower envelope line of each eigenmode according to the maximum value and the minimum value of each eigenmode includes:
fitting according to the maximum value of each eigenmode by adopting a least square method to obtain an upper envelope curve of each eigenmode; and fitting the minimum value of each eigenmode by adopting a least square method to obtain a lower envelope curve of each eigenmode.
Further, the constructing the de-intrinsic difference amplitude-frequency index of each eigenmode subsequence according to the number of zeros of each eigenmode subsequence and the upper and lower envelopes includes:
calculating the ratio of the number of the zeros of the ith subsequence of the jth eigenmode to the total number of the zeros of all subsequences of the jth eigenmode, and respectively calculating the upper envelope mean value and the lower envelope mean value of the ith subsequence of the jth eigenmode; and obtaining the absolute value of the difference between the upper envelope mean value and the lower envelope mean value, and taking the normalized result of the product of the ratio and the absolute value of the difference as the extrinsic difference amplitude-frequency index of the ith subsequence of the jth eigenmode.
Further, the constructing the temporal intrinsic-to-temporal similarity index of the subsequences from temporal intrinsic tuple differences of neighboring subsequences includes:
and calculating the absolute value of the difference value of the de-intrinsic difference amplitude-frequency index of the current subsequence and the next subsequence in the same eigenmode, and taking the reciprocal of the mean value of the absolute value of the difference values in all eigenmodes as the similarity index between the moment eigens of the current subsequence.
Further, the constructing the time intrinsic tuple heterogeneity of each sub-sequence according to the element variance in the time intrinsic tuple of each sub-sequence and the similarity index between the time intrinsic tuples comprises: the ratio of the variance of the elements in the moment intrinsic tuple of each subsequence to the similarity index between moment intrinsic tuples is taken as the moment intrinsic tuple heterogeneity of each subsequence.
Further, the obtaining the high-side current potential noise figure of the subsequence according to the subsequence and the time intrinsic tuple heterogeneity of each subsequence in the preset neighborhood window comprises the following steps:
for each subsequence;
dividing a neighborhood window with a preset size by taking the subsequence as the center, calculating the time intrinsic tuple heterogeneity average value of all the subsequences in the neighborhood window, calculating the difference absolute value of the time intrinsic tuple heterogeneity of each subsequence in the neighborhood window and the average value, and taking the average value of the difference absolute values of all the subsequences in the neighborhood window as the high-side current potential noise index of the subsequence corresponding to the center of the neighborhood window.
Further, the obtaining the window size of the moving average denoising algorithm when denoising each subsequence according to the high-side current potential noise figure of each subsequence specifically includes:
and taking the high-side current potential noise index of each subsequence as an index of an exponential function taking a natural constant as a base, and taking the sum of 2 times and 1 of the calculation result of the exponential function which is rounded upwards as the window size when denoising each subsequence.
Further, the method for detecting the denoised high-side current sequence by combining the local outlier factor detection algorithm comprises the following steps:
inputting the denoised high-side current sequence into a local outlier factor detection algorithm, outputting LOF values of all current data in the denoised high-side current sequence, and taking the current with the LOF value higher than an abnormal threshold value as abnormal current data; otherwise, the normal current data are obtained.
In a second aspect, an embodiment of the present invention further provides a high-side current detection system of a switching power supply, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the steps of the high-side current detection method of any one of the foregoing when executing the computer program.
The invention has at least the following beneficial effects:
according to the invention, the original high-side current is subjected to empirical mode decomposition, noise is suppressed, and finally, the real-time accurate detection of the switching power supply is realized. Firstly, carrying out empirical mode decomposition on an original high-side current, converting a non-stationary sequence signal into stationary modal components, and calculating an extrinsic difference amplitude-frequency index of the non-stationary sequence signal so as to be used for comparison calculation among different modalities; then calculating a time intrinsic tuple heterogeneity sequence according to the distribution relation among all modes, and calculating a high-side current potential noise figure according to the time intrinsic tuple heterogeneity sequence; and finally, denoising the original signal through the potential noise figure of the high-side current to obtain a more accurate high-side current signal of the switching power supply, and improving the high-side current detection precision of the switching power supply. According to the invention, the potential noise position is confirmed by empirical mode decomposition of the original high-side power supply signal, and the denoising algorithm can be adaptively adjusted according to the potential noise index of the high-side current, so that the high-side current running condition of the switching power supply can be accurately detected finally.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of steps of a method for detecting a high-side current of a switching power supply according to an embodiment of the present invention;
fig. 2 is a diagram of high side current noise figure extraction.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to specific implementation, structure, characteristics and effects of a high-side current detection method and system for a switching power supply according to the invention, which are provided by the invention, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a high-side current detection method and a system for a switching power supply provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for detecting a high-side current of a switching power supply according to an embodiment of the invention is shown, the method includes the following steps:
step S001, obtaining a high-side measured current signal of the switching power supply.
In general, in order to detect the current of a switching power supply, a current sensor is required to be connected to the high side to acquire a real-time current signal when the power supply is operated. To avoid errors due to switching additional loads into the circuit, the current signal changes are estimated here by a contactless hall effect current sensor, which receives the changes in the electromagnetic field. The collected current signal is sampled at 200Hz, and the sampled result is recorded as the original high-side current sequence。
Thus, the original high-side current data can be obtained according to the method of the embodiment and used as the basic data for subsequent analysis.
Step S002, decomposing according to the obtained current signals, analyzing the subsequences of each decomposed intrinsic mode, and constructing a high-side current potential noise figure.
In the detection of the current signal, the signal is affected by environmental pilot noise during acquisition, i.e. not all current anomaly fluctuations are caused by faults. Therefore, the current signal needs to be deeply analyzed. For the original high-side current sequenceIt is first divided into sub-sequences that are time identical, i.e. data for each second is taken as one sub-sequence.
First, for the original high-side current sequenceEmpirical Mode Decomposition (EMD) was performed and its mode distribution singular degree sequence was analyzed>。
For the original high-side current sequenceTypically nonlinear and non-stationary signals, so here an empirical mode decomposition method is used to decompose the signal into a plurality of eigenmode functions (IMFs) and a residual term to achieve multi-scale analysis and vibration mode extraction of the signal.
Wherein the input of the empirical mode decomposition is a time-series signal which is adaptively decomposed into a plurality of eigenmodes according to the data characteristics of the signal itself, denoted asComprising n eigenmodes and a residual term +.>. It should be noted that, the empirical mode decomposition algorithm and the specific methodAs is known in the art, the modal distributions of the signal should be uniform, i.e. they have similar frequencies and locations, typically in the absence of the effects of the noise signal.
First, for each eigenmode, the present embodiment will calculate the de-eigen difference amplitude-frequency index of the eigenmode。
Since the frequency concentration and the location of the amplitude transitions should be uniform among the different eigenmodes, the references of the frequency and amplitude are different among the different eigenmodes. Thus, the fundamental differences from the different eigenvectors need to be removed. The number of the extreme points of each eigenmode is identical with or different from the number of the crossing zero points by 1 when the empirical mode decomposition is carried out, namely, the data points of the eigenmodes are alternately distributed on two sides of the coordinate axis, so that the number of the zero points can be used as an index of the frequency analysis of the eigenmodes.
Then, fitting an upper envelope L1 by a least square method according to the maximum value point of the eigenmode, and fitting a lower envelope L2 by a least square method according to the minimum value point of the eigenmode. In order to improve the detection efficiency, the embodiment divides each intrinsic mode into a plurality of subsequences, and then analyzes based on each subsequence, thereby improving the detection speed and facilitating the analysis of the time sequence status of each intrinsic mode. It should be noted that, the number of the sub-sequences of the eigen mode division may be set by the practitioner according to the actual situation, which is not particularly limited in this embodiment.
Finally, the evaluation index is represented according to the number of the zeros of the eigenmodes and the amplitude index is represented by the upper envelope curve and the lower envelope curve of the eigenmodes. The following is shown:
，
wherein,represents the j-th eigenmode->The de-intrinsic differential amplitude-frequency index of the ith subsequence;representing normalizing the data; />Represents the j-th eigenmode->Zero number of the ith sub-sequence; />Represents the j-th eigenmode->Is the zero point total number of (2); />Represents the j-th eigenmode->An upper envelope mean of the ith subsequence; />Represents the j-th eigenmode->The lower envelope mean of the ith sub-sequence.
When the number of zero points in the same subsequence is larger, the frequency of the subsequence in the current eigenmode is larger, otherwise, the frequency of the subsequence is smaller; when the average difference between the upper envelope and the lower envelope of the current eigenmode at the time of the subsequence is larger, the larger the amplitude at the time is indicated, otherwise, the smaller the amplitude at the time is indicated.
Thus, the de-eigenvalue difference amplitude-frequency index sequence of the jth eigenvector is obtainedThe same applies to obtain the de-eigenvalue difference amplitude-frequency index sequence of other n-1 eigenvectors.
To this end, the above procedure according to the present embodiment is performed on the original high-side current sequenceAnd (5) performing empirical mode decomposition, and obtaining an extrinsic difference amplitude-frequency index sequence of each intrinsic mode. Since the locations of the frequency concentration and amplitude transitions should be identical in different eigenmodes in the absence of noise, the temporal eigen-tuples of each sub-sequence are constructed at the time of each sub-sequence as follows:
，
wherein,a temporal intrinsic tuple representing an ith subsequence; />Indicating the de-eigenvalue differential amplitude-frequency index of the ith subsequence of the nth eigenvalue.
Wherein the inner elements of the moment intrinsic tuple of the ith sub-sequence represent the de-intrinsic difference amplitude-frequency indices of different eigenmodes at the same moment. I.e. normally the smaller the noise the smaller the time instant intrinsic tuple internal element variance should be, whereas the larger the time instant intrinsic tuple internal element variance should be.
Meanwhile, when noise is smaller, the temporal intrinsic tuple similarity of the adjacent two sub-sequences should be larger, whereas the temporal intrinsic tuple similarity of the adjacent two sub-sequences should be smaller. And construct a similarity index between the intrinsic moments of the modes based on the obtained resultsThe method is characterized by comprising the following steps:
，
wherein,an inter-temporal intrinsic similarity index representing the ith subsequence; n represents the current sequence for the original high sideThe number of the intrinsic modes obtained by empirical mode decomposition; />Indicating the de-eigenvalue difference amplitude-frequency index of the ith subsequence of the jth eigenvalue; />An eigenvalue difference amplitude-frequency index representing the jth eigenmode ith-1 subsequence,/-)>To avoid the parameter with zero denominator, the practitioner can set himself, in this embodiment to +.>。
Finally, constructing the time intrinsic tuple heterogeneity according to the time intrinsic tuple internal element variance and the similarity index between the time intrinsic tupleThe method is characterized by comprising the following steps:
，
wherein,indicating temporal intrinsic tuple heterogeneity of the ith subsequence; />The moment intrinsic tuple representing the ith sub-sequenceThe standard deviation between the internal elements is used for representing the similarity of different eigenmodes at the moment of the ith subsequence,an intrinsic-to-intrinsic similarity index indicating the instant of the ith subsequence,>to avoid the parameter with zero denominator, the practitioner can set himself, in this embodiment to +.>。
When the standard deviation between the internal elements of the intrinsic tuple at the moment of the ith sub-sequence is larger, the difference of different intrinsic modes between the moments of the sub-sequence is larger, the sub-sequence moment is more likely to contain noise, otherwise, the difference of different intrinsic modes between the moments of the sub-sequence is smaller, the sub-sequence moment is less likely to contain noise; similarly, the intrinsic-to-intrinsic similarity index at the time of the ith subsequenceThe larger the time instant intrinsic tuple similarity between different time instants is indicated the less likely the sub-sequence will contain noise, whereas the lower the time instant intrinsic tuple similarity between different time instants is indicated the more likely the sub-sequence will contain noise.
To this end, through the original high-side current sequenceThe time intrinsic tuple heterogeneity sequence of each subsequence can be obtained through empirical mode decomposition and the above process.
Finally, the embodiment calculates the potential noise figure of the high-side current according to the time intrinsic tuple heterogeneity sequence. Consider the temporal intrinsic tuple heterogeneity with neighboring subsequences if the temporal intrinsic tuple heterogeneity of the current subsequence is highThe greater the difference, the more likely the sub-sequence is noise, while the higher the temporal intrinsic tuple heterogeneity of the current sub-sequence and the smaller the difference from the temporal intrinsic tuple heterogeneity of the adjacent sub-sequence, the more likely the signal fluctuations due to faults, independent of noise. Thus, by the above steps, a high side current potential noise figure is constructed>The high side current noise figure extraction schematic is shown in fig. 2. For each subsequence, dividing a neighborhood window with the size of M by taking the current subsequence as the center, and analyzing the high-side current potential noise figure of the current subsequence according to the time intrinsic tuple heterogeneity of each subsequence in the neighborhood window, wherein the specific expression is as follows:
，
wherein,a high side current potential noise figure representing the ith sub-sequence; m represents the size of a neighborhood window of the ith subsequence, and the value is 9; />Indicating temporal intrinsic tuple heterogeneity of a kth subsequence within a neighborhood window of the ith subsequence; />Representing the average of the intrinsic tuple heterogeneity at all sub-sequence moments within the neighborhood window of the ith sub-sequence.
When the difference between the instant intrinsic tuple heterogeneity of the ith sub-sequence and the instant intrinsic tuple heterogeneity of other sub-sequences in the surrounding neighborhood window is larger, the sub-sequence is more likely to contain noise, otherwise, the sub-sequence is less likely to contain noise.
Step S003, according to the potential noise figure of the high-side current, self-adaptive moving average denoising is carried out on the original high-side current sequence, so that a more accurate current signal is obtained, and the denoised current data is detected by combining a local outlier factor detection algorithm.
So far, according to step S002, the high-side current potential noise figure sequence of the original high-side current can be obtained, and then the moving average denoising algorithm is adopted to denoise the original high-side current sequence based on the high-side current potential noise figure sequence. The moving average is a common method for denoising sequence data, which reduces noise by calculating an average value in a data window, specifically: and determining the size of a window, initializing a moving average sequence, calculating an average value by a sliding window, and outputting the moving average sequence. It should be noted that, the specific moving average denoising algorithm and process are known techniques, and are not described herein.
Here by high side current potential noise figure sequenceThe window sizes of different subsequences in moving average denoising are controlled as follows:
，
wherein,a window size representing the ith subsequence; />High side current noise figure representing the ith sub-sequence,/->To round the symbol up.
The process of reconstructing the decomposed signal to obtain the original signal is the prior art, and the embodiment does not limit this, and an implementer can reconstruct and restore each sub-sequence after denoising by himself. In this embodiment, data reconstruction by using SVMD is adopted, the SVMD algorithm is a process of recombining the decomposed modal functions into the original signal, and the reconstructed signal can be obtained by weighting and adding each modal function, and the specific SVMD data reconstruction process is in the prior art and will not be described herein.
And finally, extracting abnormal current data of the denoised high-side current sequence by combining the denoised high-side current sequence with an abnormal detection algorithm so as to realize accurate detection of the high-side current of the switching power supply. The abnormal detection algorithm is a local outlier factor detection algorithm, namely an LOF algorithm, an LOF algorithm is adopted to calculate the LOF value of each current data of the denoised high-side current sequence, an abnormal threshold is set, when the LOF value of the current data is higher than the abnormal threshold, the corresponding current data is abnormal current data, otherwise, the abnormal current data is normal current data, and the accurate detection of the high-side current of the switching power supply is realized based on the abnormal current data, so that related management staff can accurately know the operation condition of the high-side current of the switching power supply, and relatively accurate reference information is provided. The abnormality threshold setting practitioner can set the abnormality threshold to 0.75 in this embodiment by himself.
Based on the same inventive concept as the method, the embodiment of the invention also provides a high-side current detection system of a switching power supply, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor realizes the steps of any one of the method of the high-side current detection method of the switching power supply when executing the computer program.
In summary, according to the embodiment of the invention, the original high-side current is subjected to empirical mode decomposition, noise is suppressed, and finally real-time detection of the switching power supply is realized. The original high-side current is subjected to empirical mode decomposition, a non-stationary sequence signal is converted into stationary modal components, and the extrinsic difference amplitude-frequency index of the non-stationary sequence signal is calculated so as to be used for comparison calculation among different modalities; calculating a time intrinsic tuple heterogeneity sequence according to the distribution relation among all modes, and calculating a high-side current potential noise figure according to the time intrinsic tuple heterogeneity sequence; finally, denoising the original signal through the potential noise figure of the high-side current to obtain a more accurate high-side current signal of the switching power supply, and combining with an LOF anomaly detection algorithm to accurately detect the high-side current of the switching power supply;
according to the embodiment of the invention, the potential noise position is confirmed by empirical mode decomposition of the original high-side power supply signal, and the denoising algorithm can be adaptively adjusted according to the potential noise index of the high-side current, so that accurate extraction of current information is finally realized, and the current detection precision is improved.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (5)
1. The high-side current detection method for the switching power supply is characterized by comprising the following steps of:
collecting high-side current signals of a switching power supply to form an original high-side current sequence;
obtaining each intrinsic mode according to the original high-side current sequence; obtaining upper and lower envelope curves of each eigenmode according to the maximum and the minimum values of each eigenmode; dividing each eigenmode according to a time sequence to obtain each subsequence of each eigenmode, and constructing an extrinsic difference amplitude-frequency index of each eigenmode subsequence according to the number of zeros of each eigenmode subsequence and upper and lower envelope curves; the eigenvalue frequency index of the difference of the extrinsic of each subsequence in all eigenvectors is formed into the moment eigenvector of each subsequence; constructing a time intrinsic similarity index of the subsequences according to the time intrinsic tuple difference of the adjacent subsequences; constructing the time intrinsic tuple heterogeneity of each subsequence according to the element variance in the time intrinsic tuple of each subsequence and the similarity index between the time intrinsic tuples; obtaining a high-side current potential noise index of the subsequence according to the subsequence and the time intrinsic tuple heterogeneity of each subsequence in a preset neighborhood window;
acquiring window sizes of each subsequence when the moving average denoising algorithm denoises the subsequences according to the high-side current potential noise indexes of each subsequence, denoising and reconstructing each subsequence to obtain a denoised high-side current sequence, and detecting the denoised high-side current sequence by combining a local outlier factor detection algorithm to finish detection of the high-side current of the switching power supply;
the obtaining each intrinsic mode according to the original high-side current sequence comprises the following steps: decomposing the original high-side current sequence by adopting an empirical mode decomposition algorithm, and outputting each intrinsic mode of the original high-side current sequence;
the method for constructing the de-intrinsic difference amplitude-frequency index of each intrinsic mode subsequence according to the number of the zeros of each intrinsic mode subsequence and the upper envelope line and the lower envelope line comprises the following steps:
calculating the ratio of the number of the zeros of the ith subsequence of the jth eigenmode to the total number of the zeros of all subsequences of the jth eigenmode, and respectively calculating the upper envelope mean value and the lower envelope mean value of the ith subsequence of the jth eigenmode; obtaining the absolute value of the difference between the upper envelope mean value and the lower envelope mean value, and taking the normalized result of the product of the ratio and the absolute value of the difference as the extrinsic difference amplitude-frequency index of the ith subsequence of the jth eigenmode;
the constructing the temporal intrinsic-to-temporal similarity index of the subsequences based on the temporal intrinsic tuple differences of adjacent subsequences includes:
calculating the difference absolute value of the extrinsic difference amplitude-frequency index of the current subsequence and the next subsequence in the same eigenmode, and taking the reciprocal of the mean value of the difference absolute values in all eigenmodes as the similarity index between the moment eigens of the current subsequence;
the method for obtaining the high-side current potential noise index of the subsequence according to the subsequence and the time intrinsic tuple heterogeneity of each subsequence in the preset neighborhood window comprises the following steps:
for each subsequence;
dividing a neighborhood window with a preset size by taking a subsequence as a center, calculating the time intrinsic tuple heterogeneity average value of all subsequences in the neighborhood window, calculating the difference absolute value of the time intrinsic tuple heterogeneity of each subsequence in the neighborhood window and the average value, and taking the average value of the difference absolute values of all subsequences in the neighborhood window as the high-side current potential noise index of the subsequence corresponding to the center of the neighborhood window;
the method for obtaining the window size of each subsequence when denoising the subsequence by using the moving average denoising algorithm according to the high-side current potential noise figure of each subsequence specifically comprises the following steps:
and taking the high-side current potential noise index of each subsequence as an index of an exponential function taking a natural constant as a base, and taking the sum of 2 times and 1 of the calculation result of the exponential function which is rounded upwards as the window size when denoising each subsequence.
2. The method for detecting high-side current of switching power supply according to claim 1, wherein said obtaining upper and lower envelopes of each eigenmode according to the maximum and minimum values of each eigenmode comprises:
fitting according to the maximum value of each eigenmode by adopting a least square method to obtain an upper envelope curve of each eigenmode; and fitting the minimum value of each eigenmode by adopting a least square method to obtain a lower envelope curve of each eigenmode.
3. The method for detecting high-side current of switching power supply as claimed in claim 1, wherein said constructing the temporal intrinsic tuple heterogeneity of each sub-sequence according to the intra-temporal intrinsic tuple element variance and the temporal intrinsic-to-temporal similarity index of each sub-sequence comprises: the ratio of the variance of the elements in the moment intrinsic tuple of each subsequence to the similarity index between moment intrinsic tuples is taken as the moment intrinsic tuple heterogeneity of each subsequence.
4. The method for detecting the high-side current of the switching power supply according to claim 1, wherein the method for detecting the denoised high-side current sequence by combining the local outlier factor detection algorithm comprises the following steps:
inputting the denoised high-side current sequence into a local outlier factor detection algorithm, outputting LOF values of all current data in the denoised high-side current sequence, and taking the current with the LOF value higher than an abnormal threshold value as abnormal current data; otherwise, the normal current data are obtained.
5. A switching power supply high side current detection system comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1-4 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410166774.5A CN117713505B (en) | 2024-02-06 | 2024-02-06 | High-side current detection method and system for switching power supply |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410166774.5A CN117713505B (en) | 2024-02-06 | 2024-02-06 | High-side current detection method and system for switching power supply |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117713505A CN117713505A (en) | 2024-03-15 |
| CN117713505B true CN117713505B (en) | 2024-04-05 |
Family
ID=90157466
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410166774.5A Active CN117713505B (en) | 2024-02-06 | 2024-02-06 | High-side current detection method and system for switching power supply |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117713505B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011257362A (en) * | 2010-06-11 | 2011-12-22 | Takada Corp | Abnormality diagnosis method of rotary machine system |
| CN111144644A (en) * | 2019-12-24 | 2020-05-12 | 淮阴工学院 | Short-term wind speed prediction method based on variation variance Gaussian process regression |
| CN114169237A (en) * | 2021-11-30 | 2022-03-11 | 南昌大学 | Power cable joint temperature abnormity early warning method combining EEMD-LSTM and isolated forest algorithm |
| CN116993002A (en) * | 2023-06-29 | 2023-11-03 | 国网辽宁省电力有限公司大连供电公司 | Load electric quantity combination prediction method and device for realizing same |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106208828B (en) * | 2015-05-27 | 2020-09-29 | 罗姆股份有限公司 | Motor drive device, motor drive IC, and cooling device and electronic apparatus using the same |
-
2024
- 2024-02-06 CN CN202410166774.5A patent/CN117713505B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011257362A (en) * | 2010-06-11 | 2011-12-22 | Takada Corp | Abnormality diagnosis method of rotary machine system |
| CN111144644A (en) * | 2019-12-24 | 2020-05-12 | 淮阴工学院 | Short-term wind speed prediction method based on variation variance Gaussian process regression |
| CN114169237A (en) * | 2021-11-30 | 2022-03-11 | 南昌大学 | Power cable joint temperature abnormity early warning method combining EEMD-LSTM and isolated forest algorithm |
| CN116993002A (en) * | 2023-06-29 | 2023-11-03 | 国网辽宁省电力有限公司大连供电公司 | Load electric quantity combination prediction method and device for realizing same |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117713505A (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Übeyli | ECG beats classification using multiclass support vector machines with error correcting output codes | |
| Kim et al. | ECG signal compression and classification algorithm with quad level vector for ECG holter system | |
| Li et al. | Damage localization of stacker’s track based on EEMD-EMD and DBSCAN cluster algorithms | |
| CN110705456A (en) | Micro motor abnormity detection method based on transfer learning | |
| CN112766342A (en) | Abnormity detection method for electrical equipment | |
| Feng et al. | Design and implementation of a self-validating pressure sensor | |
| CN116701822B (en) | Load electric energy metering method based on sparrow search algorithm and variation modal decomposition | |
| CN117421610B (en) | Data anomaly analysis method for electric energy meter running state early warning | |
| CN116933044A (en) | Intelligent processing method and system for power supply data | |
| CN111915449A (en) | VMD and OMP-based power load data dimension reduction reconstruction processing method | |
| CN117633696B (en) | Hydraulic engineering construction monitoring method and system based on BIM model | |
| CN111260776A (en) | Three-dimensional shape reconstruction method for adaptive normal analysis | |
| CN117713505B (en) | High-side current detection method and system for switching power supply | |
| Li et al. | A fault diagnosis method for rotating machinery with semi-supervised graph convolutional network and images converted from vibration signals | |
| CN104352234A (en) | Detection method for peak singular point of physiologic electric signal | |
| CN108334822B (en) | Kalman and modified wavelet transform filtering method based on electric vehicle charging nonlinear load characteristics | |
| CN116628620A (en) | Non-invasive load identification calculation method | |
| Song et al. | Research on rolling bearing fault diagnosis method based on improved LMD and CMWPE | |
| CN115828144A (en) | Signal sparse representation and fusion detection method, storage medium and electronic device | |
| CN115508618A (en) | Quasi-synchronous harmonic analysis device and method based on time domain Hermite interpolation | |
| CN113469502B (en) | Dictionary learning-based differential grouping non-invasive load decomposition method | |
| Pan et al. | Fast fault diagnosis method of rolling bearings based on compression features in multi-sensor redundant observation environment | |
| Deng et al. | A fractional hilbert transform order optimization algorithm based DE for bearing health monitoring | |
| Su et al. | A nonparametric derivative-based method for R wave detection in ECG | |
| Fan et al. | Voltage disturbance signals identification based on ILMD and neural network |
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 |