CN110133396B - Method for detecting household electrical appliance switch event based on binary recursive SVD decomposition - Google Patents
Method for detecting household electrical appliance switch event based on binary recursive SVD decomposition Download PDFInfo
- Publication number
- CN110133396B CN110133396B CN201910280687.1A CN201910280687A CN110133396B CN 110133396 B CN110133396 B CN 110133396B CN 201910280687 A CN201910280687 A CN 201910280687A CN 110133396 B CN110133396 B CN 110133396B
- Authority
- CN
- China
- Prior art keywords
- svd
- decomposition
- active power
- binary recursive
- residential
- 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
Images
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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R21/00—Arrangements for measuring electric power or power factor
- G01R21/001—Measuring real or reactive component; Measuring apparent energy
- G01R21/002—Measuring real component
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
Abstract
The invention relates to a method for detecting a switching event of a residential appliance based on binary recursive SVD (singular value decomposition), which comprises the following steps of: (1) collecting data; (2) performing binary recursive SVD on the total active power data of the resident electrical appliances, and determining the optimal decomposition layer number by calculating singular entropy; (3) filtering by using a hard threshold filtering function to obtain the switching time of the household electrical appliance; the method is used for detecting the household electrical appliance switch events based on the binary recursive SVD decomposition, and has the advantages of simple and easy operation, high calculation speed, low missing detection rate and false detection rate, high detection precision and stable detection result.
Description
Technical Field
The invention relates to the technical field of non-invasive load monitoring, in particular to a residential appliance switch event detection method based on binary recursive SVD decomposition.
Background
With the advancement of urbanization and the adjustment of economic structures in China, the proportion of the domestic electricity consumption of residents to the total electricity consumption of the whole society is larger and larger. The electricity consumption data of residents can help users to adjust electricity consumption behaviors and actively participate in demand response, and can also help power companies to reasonably arrange dispatching, peak clipping and valley filling. The load monitoring technology is an important way for obtaining the electricity consumption data of residents and can be divided into an invasive type and a non-invasive type, and the traditional invasive type monitoring technology needs to install an acquisition device and a sensing device for each piece of electric equipment, so that a large amount of manpower and material resources are consumed. Professor Hart in the 80 th 20 th century proposed a non-invasive load monitoring (NILM) technique, which can obtain the power consumption information of each electric appliance in a house by load decomposition by collecting the total load information at the user's home-entry end without installing a large amount of monitoring equipment. With the improvement of the intelligent level, the technology gradually becomes a domestic research hotspot.
The non-invasive load monitoring technology comprises three links of load switch event detection, characteristic extraction and load identification, and the accurate detection of the switch event is an important prerequisite and basis for realizing the non-invasive load monitoring. At present, the detection method of the switch event of the resident electrical appliance mainly comprises a nonparametric bilateral sliding window CUSUM variable point detection method and a wavelet decomposition method. The sliding window CUSUM point change detection method can detect the occurrence time of an electric appliance switch event by detecting the change of active power in the sliding window, but the method is limited by a set power sequence sliding window and a threshold value, and is easy to have missed detection and false detection. The wavelet decomposition method is to detect the position of a mutation point in an active power signal by performing wavelet decomposition on the power signal, but the method is difficult to select a proper wavelet base, the detection position is easy to deviate, and the detection precision is not high.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a method for detecting the switching events of the residential electrical appliances based on binary recursive SVD (singular value decomposition), so that the accuracy of detecting the switching events of the residential electrical appliances in the non-invasive load monitoring technology is improved.
The purpose of the invention can be realized by the following technical scheme:
a method for detecting a switching event of a residential appliance based on binary recursive SVD decomposition comprises the following steps:
step 1: data acquisition: collecting power consumption information data of the resident families at a power household entry point and obtaining total active power data of the resident families;
step 2: and decomposing total active power data by utilizing a binary recursive SVD (singular value decomposition): carrying out Hankel matrix transformation on the total active power data of the residential appliance switch events collected in the step 1, then carrying out binary recursive SVD (singular value decomposition), and calculating singular entropy to obtain the optimal decomposition layer number;
and step 3: and obtaining the switching time of the household electrical appliance by using a hard threshold filtering function: and (3) filtering the detail signals under the optimal decomposition layer number obtained in the step (2) by using a hard threshold filtering function, and realizing accurate detection of the position of a mutation point in the total active power, namely the switching time of the household electrical appliance.
Preferably, the electricity consumption information data in step 1 includes total current and voltage data of the residential electrical appliances, and the total active power data of the residential home is obtained by MATLAB and is used as a target object for detecting the switching event.
Preferably, the step 2 comprises the following substeps:
step 21: constructing a two-dimensional Hankel matrix aiming at total active power data;
step 22: decomposing a two-dimensional Hankel matrix by utilizing a binary recursive SVD (singular value decomposition) and then obtaining an approximate signal A and a detail signal D through inverse transformation;
step 23: repeating step 22 to decompose the raw total active power data into a series of approximation and detail signals;
step 24: the singular entropy deltas are calculated for the series of detail signals obtained in step 23.
Preferably, the step 3 comprises the following substeps:
step 31: taking an optimal value: if the singular entropy increment is larger than the set value, constructing a new Hankel matrix by using the approximate signal and repeatedly decomposing until the different entropy increment is smaller than the set value and the number of decomposition layers reaches the optimum;
step 32: and filtering the detail signals under the optimal decomposition layer number by adopting a hard threshold function to obtain the switching time of the residential electrical appliance.
Preferably, the two-dimensional Hankel matrix H in step 21 is:
wherein [ x ]1x2x3…xn]For a one-dimensional discrete signal corresponding to the total active power data of the residential electrical appliance, H belongs to R2×n-1。
Preferably, the first and second electrodes are formed of a metal,
preferably, the two-dimensional Hankel matrix H decomposed by the binary recursive SVD in step 22 is described by the formula:
H＝USVT
in the formula, the two-dimensional orthogonal matrix U ═ U (U)1，u2)，U∈R2×2；V＝(v1,v2,…vn-1) N-1 dimensional orthogonal matrix V ∈ R(n-1)×(n-1)(ii) a Diagonal matrix S ═ diag (λ)1,λ2),O)，S∈R2×(n-1)，λ1>>λ2，λ1And λ2First and second singular values, respectively.
Preferably, the hard threshold function in step 32 is:
where d is a hard threshold function, dkAre variables.
Compared with the prior art, the invention has the following advantages:
firstly, performing binary recursive SVD orthogonal decomposition on different hierarchical spaces, determining the optimal decomposition layer number by calculating singular entropy, and reflecting the catastrophe point characteristics in the original signal by using the detail signals of the different hierarchical spaces.
And secondly, filtering the tiny noise of the detail signal by adopting a hard threshold function, realizing accurate detection of the position of the mutation point in the total active power, and effectively eliminating the tiny noise in the detail signal, wherein the method is simple and easy to operate and has high calculation speed.
And thirdly, based on laboratory data and REDD data under different noise conditions, comparing the laboratory data and REDD data with the existing wavelet decomposition method and the unparameterized bilateral sliding CUSUM variable point detection method, proving that the technology of the invention has the advantages of low omission factor and false detection rate, small influence of noise and high detection precision.
Drawings
FIG. 1 is a flow chart of a residential appliance switch event detection method based on binary recursive SVD decomposition according to the present invention;
FIG. 2 shows an approximation matrix H in the present invention1Vector sum detail matrix H of2Wherein fig. 2(a) is an approximation matrix H1FIG. 2(b) shows a detail matrix H2The vector diagram of (1);
FIG. 3 is an exploded view of the multi-scale SVD of active power in the present invention;
fig. 4 is a comparison graph of the detection results of the switching events of the single linear constant power electrical appliance in the present invention, wherein fig. 4(a) is a graph of the active power of the water dispenser, fig. 4(b) is a graph of the detection results of 0.86s to 0.98s of the unparameterized sliding bilateral CUSUM change point detection method, fig. 4(c) is a graph of the detection results of 3.18s to 3.3s of the unparameterized sliding bilateral CUSUM change point detection method, fig. 4(d) is a graph of the detection results of 0.82s to 0.98s of the wavelet decomposition method, fig. 4(e) is a graph of the detection results of 3.16s to 3.32s of the wavelet decomposition method, fig. 4(f) is a graph of the detection results of 0.82s to 0.98s of the method of the present invention, and fig. 4(g) is a graph of the detection results of 3.14s to 3.3s of the method of the present invention;
fig. 5 is a comparison graph of the detection results of the switching event of the single nonlinear power-varying electrical appliance in the present invention, wherein fig. 5(a) is a graph of the active power of the microwave oven, fig. 5(b) is a graph of the detection results of the unparameterized sliding bilateral CUSUM variable point detection method, fig. 5(c) is a graph of the detection results of the wavelet decomposition method, and fig. 5(d) is a graph of the detection results of the method of the present invention;
fig. 6 is a comparison graph of detection results of a plurality of electrical switch events in an experiment according to an embodiment of the present invention, where fig. 6(a) is a graph of a plurality of electrical total active power curves, fig. 6(b) is a graph of detection results of a non-parametric sliding bilateral CUSUM change point detection method, fig. 6(c) is a graph of detection results of a wavelet decomposition method, and fig. 6(d) is a graph of detection results of a method according to the present invention;
fig. 7 is a comparison graph of the detection results of the REDD data set switching events in the embodiment of the present invention, in which fig. 7(a) is a graph of the REDD data power curve, fig. 7(b) is a graph of the detection results of the unparameterized sliding bilateral CUSUM change point detection method, fig. 7(c) is a graph of the detection results of the wavelet decomposition method, and fig. 7(d) is a graph of the detection results of the method of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
Examples
The general flow chart of the residential appliance switch event detection method based on the binary recursive SVD decomposition is shown in FIG. 1.
As shown in fig. 1, a method for detecting a switching event of a residential appliance based on binary recursive SVD decomposition comprises the following steps:
1) data acquisition: collecting total current and voltage data of the residential electrical appliances, and calculating the total active power of the residential electrical appliances;
2) decomposing total active power data of the residential electrical appliances by the aid of binary recursive SVD;
21) constructing a two-dimensional Hankel matrix: and (2) carrying out Hankel matrix transformation on the total active power data of the residential electrical appliances collected in the step 1) to construct a two-dimensional Hankel matrix H.
The total active power data of the household electrical appliance is a one-dimensional discrete signal X ═ X1x2x3…xn]Constructing a space matrix with the row number of 2 through Hankel matrix transformation:
H∈R2×n-1the H matrix is characterized in that the second row of data lags behind the first row by only one data point, so that the two rows of data of the H matrix are highly correlated. After SVD decomposition, the H matrix has and only has two singular values, and is characterized in that the first singular value is larger, and the second singular value is largerThe singular values are small.
H＝USVT (2)
Wherein U is (U)1,u2)，U∈R2×2；V＝(v1,v2,…vn-1)，V∈R(n-1)×(n-1)；S＝(diag(λ1,λ2),O)，S∈R2×(n-1)，λ1>>λ2。
22) SVD decomposes the H matrix: according to the formula
Solving an approximation matrix H1And a detail matrix H2。
In the formula ui∈R2×1，vi∈R(n-1)×1And i is 1 and 2. Let Hi＝λiuivi TThen H isi∈R2×(n-1)，HiThere is a similar matrix structure as H. H1Corresponding to larger singular values, called approximation matrix; h2Corresponding to the smaller singular values, called detail matrix. H1And H2The one-dimensional vectors obtained through the Hankel inverse transformation are called approximate signals A and detail signals D, and respectively represent the main body overview and detail features of the original signals.
In the process of solving the approximate signal A, H1With two row vectors, each being lambda1u1,1v1 TAnd λ1u1,2v1 TWherein u is1,1And u1,2Is u1Two coordinates of (a). Is provided with La1And La2The sub-vectors of the two line vectors, respectively, both represent the data a in the approximation signal A1,2,a1,3,…a1,n-1FIG. 2(a) is an approximate matrix H, as shown in FIG. 21FIG. 2(b) shows a detail matrix H2The vector of (2).
23) Solving approximationsSignal a and detail signal D: l isa1And La2Not exactly equal, to obtain a completely approximate signal of information, according to the formula a ═ a1,1，(La1+La2)/2,a1,n) And formula D ═ D1,1,(Ld2+Ld2)/2,d1,n) From H1、H2The related elements of the same data are averaged to form an approximate signal A and a detail signal D. The result of the j-th decomposition is AjAnd DjTo approximate the signal A after decompositionjConstructing a new matrix Hj+1And repeating the steps to decompose the original signal into a series of approximate and detail signals. The SVD decomposition of the total active power is shown in fig. 3.
24) Solving for singular entropy increment Δ Ej: detail signals obtained by the binary recursive SVD decomposition include the mutation point information of the total active power. As the number of layers of decomposition increases, the feature information contained in the detail signal increases, but the increment of the feature information decreases and the time required to detect a mutation point increases. According to the invention, the complexity of characteristic information is represented by singular entropy, and the singular entropy increment delta E under the decomposition layer number is calculated according to the formula (4) and the formula (5)j。
3) Method for obtaining switching time of household electrical appliance by using hard threshold filtering function
31) Taking an optimal value: if Δ E is judgedjIf the value is larger than the set value epsilon, a new matrix H is constructed by using the approximate signal A, and the decomposition is repeated until delta EjAnd when the decomposition layer number is less than the set value epsilon, the decomposition layer number reaches the optimum.
32) On-off time of the resident electrical appliance: and (4) carrying out binary recursive SVD (singular value decomposition) on the total active power data of the residential electrical appliances to obtain a detail signal containing small-amplitude noise. In order to accurately detect the accurate time of the event, a hard threshold function is adopted
And filtering the detail signal D under the optimal decomposition layer number, wherein the moment of the zero crossing point of the characteristic peak in the filtered curve is the switching moment of the household electrical appliance.
4) And (3) experimental verification: a test platform is set up in a laboratory, a Nissan company 8860 electric wave recorder is adopted to collect voltage and current data of 4 types of residential appliance switch events including a water dispenser, a microwave oven, an electric heater and a fan, and the sampling frequency is 2000 Hz.
The detection verification can be carried out on a single linear constant-power electric appliance switching event, a single nonlinear variable-power electric appliance switching event and a plurality of electric appliance switching events by respectively adopting a non-parametric sliding bilateral CUSUM variable point detection method, a wavelet decomposition method and the binary recursive SVD decomposition method.
FIG. 4 is a comparison of the detection results of the switching events of a single linear constant power electrical appliance
FIG. 5 is a comparison of the detection results of the switching events of a single nonlinear power-variable electrical appliance
FIG. 6 is a comparison of the detection results of multiple electrical switch events
Fig. 4(a) is an active power curve of the water dispenser, fig. 5(a) is an active power curve of the microwave oven, and fig. 6(a) is a total active power curve of a plurality of electric appliances. Fig. 4(b) and (c), (d) and (e), (f) and (g), fig. 5(b), (c) and (d) respectively correspond to the detection result diagrams under the unparameterized sliding bilateral CUSUM variable point detection method, the wavelet decomposition method and the binary recursive SVD decomposition method mentioned in the present invention.
Three experimental result graphs prove that the sliding bilateral CUSUM variable point detection method has the defects that the size of a sliding window and a threshold h are difficult to set, and has high omission factor and high false detection rate. The detection result of the wavelet decomposition method is obviously deviated, and is different under different decomposition levels. Compared with the other two methods, the method has the advantages of low missing detection rate, small number of false detections, stable detection position without deviation and high detection precision.
5) REDD database verification: the unparameterized sliding bilateral CUSUM variable point detection method, the wavelet decomposition method and the binary recursive SVD decomposition method are respectively adopted to detect the REDD data set load power data of 1 hour, the real switching event occurs 45 times, the statistical results of the three methods are shown in figure 7,
fig. 7(a) is a REDD data power curve, and fig. 7(b), (c) and (d) are detection result diagrams of the unparameterized sliding bilateral CUSUM variable point detection method, the wavelet decomposition method and the binary recursive SVD decomposition method mentioned in the present invention, respectively.
And (3) judging the detection method result by adopting three indexes, namely (1) false detection: detecting that a switching event has occurred, with virtually no event occurring; (2) omission detection: no switching event is detected, and in fact an event occurs; (3) and (3) recall ratio: the number of correctly detected switching events is a ratio of all real events; (4) mean absolute error: the average absolute error between the detected switching instant and the real instant. The statistical results are shown in table 1, and the results prove that the missed detection value of the sliding window bilateral CUSUM variable point detection is 2, the false detection value is 26, and the recall ratio is 95%; the undetected value of the wavelet decomposition method is 3, the false detection value is 6, and the recall ratio is 93 percent; the undetected value of the binary recursive SVD decomposition method is 0, the false detection value is 2, and the recall ratio reaches 100%. Compared with the other two methods, the method provided by the invention effectively improves the detection precision of the switch event of the residential electrical appliance.
TABLE 1
| Detecting the number of switches | Number of missed tests | Number of false detections | Detection accuracy (%) | Mean absolute error(s) | |
| CUSUM | 43 | 2 | 26 | 95 | 1.29 |
| Wavelet decomposition | 42 | 3 | 6 | 93 | 1.43 |
| SVD decomposition | 45 | 0 | 2 | 100 | 0.90 |
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (4)
1. A method for detecting the switching event of a residential appliance based on binary recursive SVD (singular value decomposition) is characterized by comprising the following steps of:
step 1: data acquisition: collecting power consumption information data of the resident families at a power household entry point and obtaining total active power data of the resident families;
step 2: and decomposing total active power data by utilizing a binary recursive SVD (singular value decomposition): carrying out Hankel matrix transformation on the total active power data of the residential appliance switch events collected in the step 1, then carrying out binary recursive SVD (singular value decomposition), and calculating singular entropy to obtain the optimal decomposition layer number;
and step 3: and obtaining the switching time of the household electrical appliance by using a hard threshold filtering function: filtering the detail signals under the optimal decomposition layer number obtained in the step 2 by using a hard threshold filtering function to realize accurate detection of the position of a mutation point in the total active power, namely the switching time of the household electrical appliance;
the step 2 comprises the following sub-steps:
step 21: constructing a two-dimensional Hankel matrix aiming at total active power data;
step 22: decomposing a two-dimensional Hankel matrix by utilizing a binary recursive SVD (singular value decomposition) and then obtaining an approximate signal A and a detail signal D through inverse transformation;
step 23: repeating step 22 to decompose the raw total active power data into a series of approximation and detail signals;
step 24: calculating a singular entropy increment for the series of detail signals obtained in step 23;
the two-dimensional Hankel matrix H in the step 21 is as follows:
wherein [ x ]1 x2 x3 … xn]For a one-dimensional discrete signal corresponding to the total active power data of the residential electrical appliance, H belongs to R2×n-1；
The two-dimensional Hankel matrix H decomposed by the binary recursive SVD in the step 22 has a description formula as follows:
H＝USVT
in the formula, the two-dimensional orthogonal matrix U ═ U (U)1，u2)，U∈R2×2；V＝(v1,v2,…vn-1) N-1 dimensional orthogonal matrix V ∈ R(n -1)×(n-1)(ii) a Diagonal matrix S ═ diag (λ)1,λ2),O)，S∈R2×(n-1)，λ1>>λ2，λ1And λ2First and second singular values, respectively;
the calculation formula of the singular entropy increment in the step 24 is as follows:
in the formula, λpAnd λqRespectively, singular values on the diagonal matrix, and r is the order.
2. The method for detecting the switching event of the residential electrical appliances based on the binary recursive SVD as claimed in claim 1, wherein the electricity consumption information data in step 1 comprises total current and voltage data of the residential electrical appliances, and the total active power data of the residential home is obtained by MATLAB and is used as a target object for the switching event detection.
3. The method for detecting the switching event of the residential appliance based on the binary recursive SVD decomposition as claimed in claim 1, wherein said step 3 comprises the following substeps:
step 31: taking an optimal value: if the singular entropy increment is larger than the set value, constructing a new Hankel matrix by using the approximate signal and repeatedly decomposing until the different entropy increment is smaller than the set value and the number of decomposition layers reaches the optimum;
step 32: and filtering the detail signals under the optimal decomposition layer number by adopting a hard threshold function to obtain the switching time of the residential electrical appliance.
4. The method for detecting the switching event of the household electrical appliances based on the binary recursive SVD decomposition as claimed in claim 3, wherein the hard threshold function in step 32 is:
where d is a hard threshold function, dkAre variables.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910280687.1A CN110133396B (en) | 2019-04-09 | 2019-04-09 | Method for detecting household electrical appliance switch event based on binary recursive SVD decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910280687.1A CN110133396B (en) | 2019-04-09 | 2019-04-09 | Method for detecting household electrical appliance switch event based on binary recursive SVD decomposition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110133396A CN110133396A (en) | 2019-08-16 |
| CN110133396B true CN110133396B (en) | 2021-07-20 |
Family
ID=67569327
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910280687.1A Active CN110133396B (en) | 2019-04-09 | 2019-04-09 | Method for detecting household electrical appliance switch event based on binary recursive SVD decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110133396B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101876546A (en) * | 2009-11-27 | 2010-11-03 | 北京航空航天大学 | MEMS (Micro Electronic Mechanical System) gyro data processing method based on wavelet threshold de-noising and FAR (Finite Automaton Recognizable) model |
| CN104748961A (en) * | 2015-03-30 | 2015-07-01 | 中国矿业大学 | Gear fault diagnosis method based on SVD decomposition and noise reduction and correlation EEMD entropy features |
| CN105224872A (en) * | 2015-09-30 | 2016-01-06 | 河南科技大学 | A kind of user's anomaly detection method based on neural network clustering |
| CN106706285A (en) * | 2016-11-24 | 2017-05-24 | 重庆工业职业技术学院 | Brake disc inherent frequency online detection method |
| CN107133195A (en) * | 2017-04-14 | 2017-09-05 | 大连理工大学 | A kind of Methodology of The Determination of The Order of Model of engineering structure modal idenlification |
| CN108506742A (en) * | 2018-03-10 | 2018-09-07 | 西安电子科技大学 | A kind of adaptive useful signal judgement fluid line leakage locating method |
| CN109101465A (en) * | 2018-07-13 | 2018-12-28 | 广东石油化工学院 | A kind of electric switch event detecting method and system |
| CN110443104A (en) * | 2019-03-29 | 2019-11-12 | 杭州电子科技大学 | A kind of data filtering methods applied to ultrasonic partial discharge detection |
-
2019
- 2019-04-09 CN CN201910280687.1A patent/CN110133396B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101876546A (en) * | 2009-11-27 | 2010-11-03 | 北京航空航天大学 | MEMS (Micro Electronic Mechanical System) gyro data processing method based on wavelet threshold de-noising and FAR (Finite Automaton Recognizable) model |
| CN104748961A (en) * | 2015-03-30 | 2015-07-01 | 中国矿业大学 | Gear fault diagnosis method based on SVD decomposition and noise reduction and correlation EEMD entropy features |
| CN105224872A (en) * | 2015-09-30 | 2016-01-06 | 河南科技大学 | A kind of user's anomaly detection method based on neural network clustering |
| CN106706285A (en) * | 2016-11-24 | 2017-05-24 | 重庆工业职业技术学院 | Brake disc inherent frequency online detection method |
| CN107133195A (en) * | 2017-04-14 | 2017-09-05 | 大连理工大学 | A kind of Methodology of The Determination of The Order of Model of engineering structure modal idenlification |
| CN108506742A (en) * | 2018-03-10 | 2018-09-07 | 西安电子科技大学 | A kind of adaptive useful signal judgement fluid line leakage locating method |
| CN109101465A (en) * | 2018-07-13 | 2018-12-28 | 广东石油化工学院 | A kind of electric switch event detecting method and system |
| CN110443104A (en) * | 2019-03-29 | 2019-11-12 | 杭州电子科技大学 | A kind of data filtering methods applied to ultrasonic partial discharge detection |
Non-Patent Citations (2)
| Title |
|---|
| 基于Hankel矩阵的复小波_奇异值分解法提取局部放电特征信息;唐炬，等;《中国电机工程学报》;20150405;第35卷(第7期);第1808-1817页 * |
| 基于二分递推SVD分解的居民电器开关事件检测方法;林顺富，等;《电网技术》;20200430;第44卷(第4期);第1534-1542页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110133396A (en) | 2019-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rahimpour et al. | Non-intrusive energy disaggregation using non-negative matrix factorization with sum-to-k constraint | |
| Liu et al. | A hybrid FCW-EMD and KF-BA-SVM based model for short-term load forecasting | |
| Chen et al. | Temporal and spectral feature learning with two-stream convolutional neural networks for appliance recognition in NILM | |
| CN108152630B (en) | Non-invasive IH electric cooker operation identification method based on mixing criterion | |
| CN109633301B (en) | Non-invasive electrical appliance load identification method based on quantum genetic optimization | |
| CN111553444A (en) | Load identification method based on non-invasive load terminal data | |
| CN111563827B (en) | Load decomposition method based on physical characteristics of electrical appliance and electricity utilization behaviors of residents | |
| CN108918931B (en) | Power signal self-adaptive filtering method in load decomposition | |
| CN108918930B (en) | Power signal self-adaptive reconstruction method in load decomposition | |
| Xu et al. | A new non-intrusive load monitoring algorithm based on event matching | |
| CN114037178A (en) | Non-invasive load decomposition method based on unsupervised pre-training neural network | |
| CN108918928B (en) | Power signal self-adaptive reconstruction method in load decomposition | |
| Bartman et al. | Identification of electrical appliances using their virtual description and data selection for non-intrusive load monitoring | |
| Azad et al. | Non-intrusive load monitoring (nilm) using deep neural networks: A review | |
| CN110133396B (en) | Method for detecting household electrical appliance switch event based on binary recursive SVD decomposition | |
| Hoyo-Montaño et al. | Overview of Non-Intrusive Load Monitoring: A way to energy wise consumption | |
| CN109946597A (en) | Tap switch operating status appraisal procedure based on dynamoelectric signal | |
| Han et al. | Non‐intrusive load monitoring by voltage–current trajectory enabled asymmetric deep supervised hashing | |
| Hacine-Gharbi et al. | Relevant harmonics selection based on mutual information for electrical appliances identification | |
| CN107767037A (en) | A kind of user power utilization forms analytic method | |
| Liu et al. | An Efficient Supervised Energy Disaggregation Scheme for Power Service in Smart Grid. | |
| Liu et al. | Enhanced NILM load pattern extraction via variable-length motif discovery | |
| CN112993993A (en) | Non-invasive power load perception identification method based on improved differential evolution algorithm | |
| CN109521296B (en) | Non-invasive power load identification optimization algorithm under steady-state working condition | |
| CN109194367B (en) | Power signal reconstruction method in energy decomposition |
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 |