CN114139820A - Improved modal decomposition method for non-invasive electric energy load prediction - Google Patents
Improved modal decomposition method for non-invasive electric energy load prediction Download PDFInfo
- Publication number
- CN114139820A CN114139820A CN202111485749.6A CN202111485749A CN114139820A CN 114139820 A CN114139820 A CN 114139820A CN 202111485749 A CN202111485749 A CN 202111485749A CN 114139820 A CN114139820 A CN 114139820A
- Authority
- CN
- China
- Prior art keywords
- component
- sequence
- electric energy
- energy load
- imf
- 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.)
- Pending
Links
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000011156 evaluation Methods 0.000 claims abstract description 10
- 230000006641 stabilisation Effects 0.000 claims abstract description 4
- 238000011105 stabilization Methods 0.000 claims abstract description 4
- 238000012545 processing Methods 0.000 claims description 7
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 241000228740 Procrustes Species 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
An improved modal decomposition method for non-intrusive electrical energy load prediction, comprising the steps of: (1) normalizing the collected non-invasive electric energy meter electric energy load time sequence information; (2) performing Empirical Mode Decomposition (EMD) on the electric energy load time series information; (3) according to the time scale characteristics of data, introducing white noise to carry out improved empirical mode decomposition (I-EMD); (4) calculating an Intrinsic Mode Function (IMF) component and a residual component of local characteristic signals containing different time scales; (5) carrying out stabilization treatment on the electric energy load sequence by using an I-EMD decomposition method; (6) and establishing a component zero-crossing rate evaluation index model and identifying the intrinsic mode component. The method can accurately decompose and classify the non-invasive electric energy load change characteristics so as to be beneficial to subsequent power utilization type identification.
Description
Technical Field
The invention relates to the field of power load monitoring, in particular to an improved modal decomposition method for non-intrusive type electric energy load prediction.
Background
In recent years, with the continuous development of smart grids, the lean and intelligent analysis of electrical signals has attracted more and more attention. Non-intrusive load monitoring (NILM) technology employs feature extraction and machine learning algorithms to analyze the combined voltage and current and monitor appliance usage without the need to install sub-meters. Many feature extraction methods have been proposed for non-intrusive load monitoring, such as wavelet features, voltage-current traces, current harmonics, active/reactive power, and automatically learned depth features.
Correlation studies extract signal features by matching a new set of wavelets to load types using wavelet design and procrustes analysis. However, in this method, the wavelet used according to the different load categories needs to be selected from a large number of wavelets, so that the use of the wavelet in the unknown appliance scene is limited.
The V-I track class characteristics extract different characteristics from a Voltage-Current (V-I) track graph to describe the working states of different electrical appliances. The proposed V-I trajectory class features cyclic orientation, closed area, mean curve non-linearity, number of self-intersections, medium segmentation slope, left and right segmentation area, etc. However, the influence of the noise signal on the V-I track is large, and the abrupt change of various characteristics of the V-I track is easily caused.
The current harmonic characteristics are electrical appliances distinguished by extracting characteristics such as modes, mode ratios and the like from high-frequency current harmonics. However, to obtain better effect, the sampling frequency of the signal is generally required to reach 1MHZ or more, which brings great obstacles to the application of current harmonic characteristics. And performing cluster analysis on active data and reactive data of various household appliances, and taking various centers as power load characteristics of each appliance so as to train and identify. The active/reactive power characteristics make it difficult to distinguish between nearly powerful appliances.
The deep learning method needs a large amount of labeled data for supervised or semi-supervised learning, and the labeled data is often expensive and rare in an actual scene, which greatly limits the development.
Disclosure of Invention
The invention aims at the problems and provides an improved modal decomposition method for non-invasive electric energy load prediction.
The technical scheme of the invention is as follows: the method comprises the following steps:
(1) normalizing the collected non-invasive electric energy meter electric energy load time sequence information;
(2) performing empirical mode decomposition on the electric energy load time series information;
(3) according to the time scale characteristics, white noise is introduced to improve empirical mode decomposition;
(4) calculating an Intrinsic Mode Function (IMF) component and a residual component of local characteristic signals containing different time scales;
(5) carrying out stabilization treatment on the electric energy load time sequence by using an improved empirical mode decomposition method;
(6) and establishing a component zero-crossing rate evaluation index model and identifying the intrinsic mode component.
The step (1) comprises the following steps:
step (11): collecting the electric energy load time sequence of the non-invasive electric energy meter, and recording as an original signal X (t) ═ X1,X2,...,Xk),XtRepresenting the electric energy load data of the non-invasive electric energy meter acquired at the time t;
step (12): converting the original signal X (t) into a signal between (-1, 1) by carrying out normalization processing;
in the formula: xiDenotes the ith sample point data, Xmax,XminRespectively maximum and minimum values of the acquisition load time series, YiAnd normalizing the data of the current sampling point.
The step (2) comprises the following steps:
step (21): calculating all local maximum values and minimum values of the normalized original sequence Y (t), and performing interpolation fitting operation on the maximum values to obtain two smooth curves, namely an upper envelope sequence Y of the normalized original sequence Y (t)max(t) and the lower envelope sequence Ymin(t) calculating the mean m of the upper and lower envelope sequences1(t),
Step (22): subtracting the envelope mean m from the original sequence Y (t)1(t) removing low frequency to obtain new sequence
if the following conditions are met: the difference between the number of zero crossing points and the number of extreme points on the whole signal sequence is not more than 1, and the mean value of the upper envelope line and the lower envelope line of any point of the sequence is 0;
if it isIf the IMF component is not the IMF component, the IMF component is regarded as an original sequence, and the steps (21) to (23) are repeated until the condition that the IMF component is a strip is met after the k-th iteration calculationObtaining a first-order modal component signal:
step (24): c is to1(t) separating residual component from original sequence Y (t), using residual component as new sequence, repeatedly judging and separating IMF component, when residual component r isn(t) stopping decomposition when the (t) is constant or is a monotonic function to obtain n groups of sample modal components { c) with different frequencies1,c2,…cnAnd residual term rn。
The step (3) comprises the following steps:
step (31): according to the original signal X (t), forming white noise w (t) with the amplitude standard deviation being beta times of the standard deviation of the original signal sequence, wherein beta is the intensity parameter of the white noise;
step (32): adding original signal X (t) into signal noise w (t) to obtain signal X' (t);
step (33): the signal X '(t) is normalized to obtain Y' (t).
The step (4) comprises the following steps:
step (41): decomposing the signal Y' (t) by an empirical mode decomposition method to obtain each IMF component ci(t) and residual component rn(t) is:
wherein n is the number of the decomposed IMF components;
step (42): repeating the operation for N times, adding white noise sequences with the same intensity and different sequences each time, and decomposing the result into:
wherein j is the added white noise sequence in the jth group.
The step (5) comprises the following steps:
according to the characteristic that the white noise frequency spectrum mean value is zero, averaging all IMF components added with the white noise to obtain the final IMF component as follows:
the step (6) comprises the following steps:
step (61): for a discrete sequence of modal components ciIf c isi(t')·ci(t' +1) < 0, the component sequence is said to have zero crossing; traversing the component sequence, counting all zero crossing points of the component sequence, and recording as Ncross0Defining a component zero crossing rate evaluation index model as follows:
wherein, ci(t')、ci(t' +1) represents the discrete modal component ciData values at time t 'and time t' + 1; pcross0Representing the component zero crossing rate; n is a discrete modal component ciThe number of sequence points;
step (62): and taking the zero crossing rate of the sequence of 0.01 as a classification standard, dividing the components with the zero crossing rate higher than 0.01 into high-frequency components, and dividing the components with the zero crossing rate lower than 0.01 into low-frequency components.
In the work, the invention provides the improved empirical mode decomposition (I-EMD) and introduces white noise for auxiliary decomposition on the basis of the Empirical Mode Decomposition (EMD), so that the modal aliasing phenomenon is solved, the improved empirical mode decomposition (I-EMD) can be effectively applied to non-invasive load signal decomposition, the load sequence is decomposed into a plurality of modal components with strong regularity, the error of load prediction caused by sequence randomness is reduced, and the prediction result of each modal component is used for the subsequent load type identification.
The invention can accurately decompose and classify the non-invasive electric energy load change characteristics so as to be beneficial to subsequent processing.
Drawings
Figure 1 is a flow chart of the present invention,
figure 2 is a non-intrusive electrical energy loading raw sequence of an embodiment of the present invention,
figure 3 is a graph of the improved modal decomposition results of the non-intrusive loading sequence in an embodiment of the present invention,
fig. 4 is a schematic diagram of the zero crossing rate of the modal component in the embodiment of the present invention.
Detailed Description
The invention, as shown in fig. 1, comprises the following steps:
(1) normalizing the collected non-invasive electric energy meter electric energy load time sequence information;
(2) performing empirical mode decomposition on the electric energy load time series information;
(3) according to the time scale characteristics, white noise is introduced to improve empirical mode decomposition;
(4) calculating an Intrinsic Mode Function (IMF) component and a residual component of local characteristic signals containing different time scales;
(5) carrying out stabilization treatment on the electric energy load time sequence by using an improved empirical mode decomposition method;
(6) and establishing a component zero-crossing rate evaluation index model and identifying the intrinsic mode component.
In the working process, the method carries out normalization processing on the non-invasive electric energy meter electric energy load time sequence information, improves empirical mode decomposition (I-EMD), analyzes the time scale characteristics of data, solves mode aliasing, extracts intrinsic mode function IMF components and residual components, establishes a zero-crossing rate evaluation index model, and identifies intrinsic mode components. Therefore, the non-invasive electric energy load change characteristics can be accurately decomposed and classified, and subsequent processing is facilitated.
The method specifically comprises the following steps:
step (1) collecting the electric energy negative of the non-invasive electric energy meterCharge time series, denoted as original signal X (t) ═ X1,X2,...,Xk),XtRepresenting the electric energy load data of the non-invasive electric energy meter acquired at the time t; by carrying out normalization processing, improving the local part of the sequence to contain a large amount of high-frequency components and noise, and converting an original signal X (t) into a signal between (-1, 1);
in the formula: xiDenotes the ith sample point data, Xmax,XminRespectively maximum and minimum values of the acquisition load time series, YiAnd normalizing the data of the current sampling point.
Step (2) empirical Mode decomposition is carried out on the electric energy load time series information, the complicated electric energy load time series is decomposed into a finite number of eigenmode functions (IMF) through analysis and calculation, and IMF components under different modes are obtained, and the method comprises the following steps:
step (21): calculating all local maximum values and minimum values of the normalized original sequence Y (t), and performing interpolation fitting operation on the maximum values to obtain two smooth curves, namely an upper envelope sequence Y of the normalized original sequence Y (t)max(t) and the lower envelope sequence Ymin(t) calculating the mean m of the upper and lower envelope sequences1(t),
Step (22): subtracting the envelope mean m from the original sequence Y (t)1(t) removing low frequency to obtain new sequence
if the following conditions are met: the difference between the number of zero crossing points and the number of extreme points on the whole signal sequence is not more than 1, and the mean value of the upper envelope line and the lower envelope line of any point of the sequence is 0;
if it isIf the IMF component is not the IMF component, the IMF component is regarded as an original sequence, and the steps (21) to (23) are repeated until the IMF component is satisfied after the kth iteration calculation, so that a first-order modal component signal can be obtained:
Step (24): c is to1(t) separating residual component from original sequence Y (t), using residual component as new sequence, repeatedly judging and separating IMF component, when residual component r isn(t) stopping decomposition when the (t) is constant or is a monotonic function to obtain n groups of sample modal components { c) with different frequencies1,c2,…cnAnd residual term rn。
Introducing white noise to carry out improved empirical mode decomposition according to time scale characteristics, and forming white noise w (t) with amplitude standard deviation being beta times of sequence standard deviation according to an original signal X (t), wherein beta is an intensity parameter of the white noise; adding original signal X (t) into signal noise w (t) to obtain signal X' (t); the signal X '(t) is normalized to obtain Y' (t).
Step (4) decomposing the signal Y' (t) by adopting an empirical mode decomposition method to obtain each IMF component ci(t) and residual component rn(t) is:
wherein n is the number of the decomposed IMF components;
repeating the operation for N times, adding white noise sequences with the same intensity and different sequences each time, and decomposing the result into:
wherein j is the added white noise sequence in the jth group.
And (5) averaging all IMF components added with the white noise according to the characteristic that the average value of the white noise frequency spectrum is zero to obtain the final IMF component:
and (6) decomposing the normalized original electric energy load time sequence into modal components with different frequency scales by adopting an improved empirical mode decomposition method. And establishing a component zero crossing rate evaluation index model, and dividing modal components into a high-frequency component and a low-frequency component so as to introduce different load prediction models for predicting the high-frequency and low-frequency components and improve the prediction precision. The method comprises the following steps:
step (61): for a discrete sequence of modal components ciIf c isi(t')·ci(t' +1) < 0, the component sequence is said to have zero crossing. Traversing the component sequence, counting all zero crossing points of the component sequence, and recording as Ncross0Defining a component zero crossing rate evaluation index model as follows:
wherein, ci(t')、ci(t' +1) represents the discrete modal component ciData values at time t 'and time t' + 1; pcross0Representing the component zero crossing rate; n is a discrete modal component ciThe number of sequence points.
Step (62): and taking the zero crossing rate of the sequence of 0.01 as a classification standard, dividing the components with the zero crossing rate higher than 0.01 into high-frequency components, and dividing the components with the zero crossing rate lower than 0.01 into low-frequency components.
In particular, Empirical Mode Decomposition (EMD) is a major theoretical breakthrough in the research direction of linear and stationary spectral analysis based on fourier transform. The empirical mode decomposition is different from a Fourier decomposition method and a wavelet decomposition method, and the empirical mode decomposition is used for decomposing sequence signals directly according to the time scale characteristics of original data without setting any basis function on the original data and gradually decomposing the signals with different scales in the sequence into a data sequence with single characteristics. Since no basis function setting is required for the raw data, the empirical mode decomposition can theoretically decompose any type of signal.
Compared with wavelet decomposition, Empirical Mode Decomposition (EMD) carries out signal decomposition according to the time scale characteristics of data, does not need to set a priori basis function, and has obvious advantages in processing nonlinear and non-stationary sequence data.
The improved empirical mode decomposition (I-EMD) provided by the invention introduces white noise to carry out auxiliary decomposition on the basis of EMD, solves the mode aliasing phenomenon, can be effectively applied to non-invasive load signal decomposition, decomposes a load sequence into a plurality of mode components with strong regularity, reduces the error of load prediction caused by sequence randomness, and uses the prediction result of each mode component for the subsequent load type identification.
The specific application example is as follows:
the load curve data is selected as an example, and the improved empirical mode decomposition method provided by the invention is explained. The improved empirical mode decomposition (I-EMD) provided by the invention introduces white noise on the basis of EMD to carry out auxiliary decomposition, solves the modal aliasing phenomenon, decomposes a load sequence into a plurality of modal components with strong regularity, then divides each modal component into a high-frequency component and a low-frequency component by adopting the component zero-crossing rate judgment index model provided by the invention, introduces different load prediction models for the high-frequency and low-frequency components to respectively predict, improves the prediction precision, and uses the prediction result of each modal component for the subsequent load type identification.
Fig. 2 is a visual illustration of the original sequence of non-intrusive electrical energy loading. Fig. 3 is a graph of 10 modal component sequences IMF1 to IMF10 and a residual component sequence Res obtained by decomposing the normalized load sequence by using the improved empirical mode decomposition method proposed by the present invention with respect to the non-invasive electrical energy load original sequence of fig. 2, and the frequency distribution of each modal component is stable without obvious modal aliasing phenomenon.
FIG. 4 is a zero-crossing rate calculated for each modal component of FIG. 4 by using the component zero-crossing rate evaluation index model proposed by the present invention, the zero-crossing rate of IMF1-IMF6 is greater than 0.01, and therefore is classified as a high frequency component; the zero-crossing rate of the other components is less than 0.01, and the components are classified as low-frequency components. Different load prediction models can be introduced for high and low frequency components to perform prediction respectively, and prediction accuracy is improved.
Claims (7)
1. An improved modal decomposition method for non-intrusive electrical energy load prediction, comprising the steps of:
(1) normalizing the collected non-invasive electric energy meter electric energy load time sequence information;
(2) performing empirical mode decomposition on the electric energy load time series information;
(3) according to the time scale characteristics, white noise is introduced to improve empirical mode decomposition;
(4) calculating an Intrinsic Mode Function (IMF) component and a residual component of local characteristic signals containing different time scales;
(5) carrying out stabilization treatment on the electric energy load time sequence by using an improved empirical mode decomposition method;
(6) and establishing a component zero-crossing rate evaluation index model and identifying the intrinsic mode component.
2. The improved modal decomposition method for non-intrusive electrical energy load prediction as defined in claim 1, wherein step (1) comprises the steps of:
step (11): collecting the electric energy load time sequence of the non-invasive electric energy meter, and recording as an original signal X (t) ═ X1,X2,...,Xk),XtRepresenting the electric energy load data of the non-invasive electric energy meter acquired at the time t;
step (12): converting the original signal X (t) into a signal between (-1, 1) by carrying out normalization processing;
in the formula: xiDenotes the ith sample point data, Xmax,XminRespectively maximum and minimum values of the acquisition load time series, YiAnd normalizing the data of the current sampling point.
3. The improved modal decomposition method for non-intrusive electrical energy load prediction as defined in claim 1, wherein step (2) comprises the steps of:
step (21): calculating all local maximum values and minimum values of the normalized original sequence Y (t), and performing interpolation fitting operation on the maximum values to obtain two smooth curves, namely an upper envelope sequence Y of the normalized original sequence Y (t)max(t) and the lower envelope sequence Ymin(t) calculating the mean m of the upper and lower envelope sequences1(t),
Step (22): subtracting the envelope mean m from the original sequence Y (t)1(t) removing low frequency to obtain new sequence
if the following conditions are met: the difference between the number of zero crossing points and the number of extreme points on the whole signal sequence is not more than 1, and the mean value of the upper envelope line and the lower envelope line of any point of the sequence is 0;
if it isIf the IMF component is not the IMF component, the IMF component is regarded as an original sequence, and the steps (21) to (23) are repeated until the IMF component is satisfied after the kth iteration calculation, so that a first-order modal component signal can be obtained:
step (24): c is to1(t) separating residual component from original sequence Y (t), using residual component as new sequence, repeatedly judging and separating IMF component, when residual component r isn(t) stopping decomposition when the (t) is constant or is a monotonic function to obtain n groups of sample modal components { c) with different frequencies1,c2,…cnAnd residual term rn。
4. The improved modal decomposition method for non-intrusive electrical energy load prediction as defined in claim 1, wherein step (3) comprises the steps of:
step (31): according to the original signal X (t), forming white noise w (t) with the amplitude standard deviation being beta times of the standard deviation of the original signal sequence, wherein beta is the intensity parameter of the white noise;
step (32): adding original signal X (t) into signal noise w (t) to obtain signal X' (t);
step (33): the signal X '(t) is normalized to obtain Y' (t).
5. The improved modal decomposition method for non-intrusive electrical energy load prediction as defined in claim 1, wherein step (4) comprises the steps of:
step (41): decomposing the signal Y' (t) by an empirical mode decomposition method to obtain each IMF component ci(t) and residual component rn(t) is:
wherein n is the number of the decomposed IMF components;
step (42): repeating the operation for N times, adding white noise sequences with the same intensity and different sequences each time, and decomposing the result into:
wherein j is the added white noise sequence in the jth group.
6. The improved modal decomposition method for non-intrusive electrical energy load prediction as defined in claim 1, wherein step (5) comprises the steps of:
according to the characteristic that the white noise frequency spectrum mean value is zero, averaging all IMF components added with the white noise to obtain the final IMF component as follows:
7. the improved modal decomposition method for non-intrusive electrical energy load prediction as defined in claim 1, wherein step (6) comprises the steps of:
step (61): for a discrete sequence of modal components ciIf c isi(t')·ci(t' +1) < 0, the component sequence is said to have zero crossing; traversing the component sequence, counting all zero crossing points of the component sequence, and recording as Ncross0Defining a component zero crossing rate evaluation index model as follows:
wherein, ci(t')、ci(t' +1) represents the discrete modal component ciData values at time t 'and time t' + 1; pcross0Representing the component zero crossing rate; n is a discrete modal component ciThe number of sequence points;
step (62): and taking the zero crossing rate of the sequence of 0.01 as a classification standard, dividing the components with the zero crossing rate higher than 0.01 into high-frequency components, and dividing the components with the zero crossing rate lower than 0.01 into low-frequency components.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111485749.6A CN114139820A (en) | 2021-12-07 | 2021-12-07 | Improved modal decomposition method for non-invasive electric energy load prediction |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111485749.6A CN114139820A (en) | 2021-12-07 | 2021-12-07 | Improved modal decomposition method for non-invasive electric energy load prediction |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114139820A true CN114139820A (en) | 2022-03-04 |
Family
ID=80384647
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111485749.6A Pending CN114139820A (en) | 2021-12-07 | 2021-12-07 | Improved modal decomposition method for non-invasive electric energy load prediction |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114139820A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116703183A (en) * | 2023-08-04 | 2023-09-05 | 国网吉林省电力有限公司经济技术研究院 | Novel power distribution network system power load assessment method |
CN116755641A (en) * | 2023-08-22 | 2023-09-15 | 山东凌远机电科技有限公司 | Distribution box operation data optimization acquisition and storage method |
-
2021
- 2021-12-07 CN CN202111485749.6A patent/CN114139820A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116703183A (en) * | 2023-08-04 | 2023-09-05 | 国网吉林省电力有限公司经济技术研究院 | Novel power distribution network system power load assessment method |
CN116703183B (en) * | 2023-08-04 | 2023-09-29 | 国网吉林省电力有限公司经济技术研究院 | Novel power distribution network system power load assessment method |
CN116755641A (en) * | 2023-08-22 | 2023-09-15 | 山东凌远机电科技有限公司 | Distribution box operation data optimization acquisition and storage method |
CN116755641B (en) * | 2023-08-22 | 2023-10-24 | 山东凌远机电科技有限公司 | Distribution box operation data optimization acquisition and storage method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Munshi et al. | Unsupervised nonintrusive extraction of electrical vehicle charging load patterns | |
CN111564842B (en) | Method for statistical estimation of medium-and-long-term energy consumption in non-invasive electric load monitoring | |
CN114139820A (en) | Improved modal decomposition method for non-invasive electric energy load prediction | |
CN111830347B (en) | Two-stage non-invasive load monitoring method based on event | |
CN109975673B (en) | Method for identifying fault arc at direct current side of photovoltaic microgrid | |
CN112633427B (en) | Ultrahigh-order harmonic emission signal detection method based on outlier detection | |
CN108009730A (en) | A kind of photovoltaic power station system health status analysis method | |
CN111753699A (en) | Method for improving digital metering accuracy of direct current charging pile | |
CN101995437A (en) | Method for extracting features of crack acoustic emission signal of drawing part | |
CN113011481A (en) | Electric energy meter function abnormity evaluation method and system based on decision tree algorithm | |
CN108957175B (en) | Electric energy quality disturbance identification method based on improved HHT algorithm | |
Rehman et al. | Low complexity non-intrusive load disaggregation of air conditioning unit and electric vehicle charging | |
CN101143394B (en) | Electric spark process discharging state forecasting method | |
CN110866840A (en) | Database modeling method for power load characteristic quantity training based on knowledge graph | |
CN114397569A (en) | Circuit breaker fault arc detection method based on VMD parameter optimization and sample entropy | |
CN105550450B (en) | Electric energy quality interference source characteristic harmonic modeling method | |
CN112636328B (en) | Medium-voltage distribution network equipment load state identification method | |
CN116861316B (en) | Electrical appliance monitoring method and device | |
CN117172601A (en) | Non-invasive load monitoring method based on residual total convolution neural network | |
CN115902396B (en) | Resonance detection method and device for large wind power grid-connected system | |
CN109617051B (en) | New energy power system low-frequency oscillation parameter identification method | |
CN116776181A (en) | Terminal side load identification method, medium and system based on improved fuzzy clustering | |
CN111506636A (en) | System and method for analyzing residential electricity consumption behavior based on autoregressive and neighbor algorithm | |
CN116628620A (en) | Non-invasive load identification calculation method | |
CN114662576A (en) | Non-invasive storage battery car charging detection method and system based on supervised classification |
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 |