CN113901726A - Non-invasive load decomposition method based on seq2point model - Google Patents
Non-invasive load decomposition method based on seq2point model Download PDFInfo
- Publication number
- CN113901726A CN113901726A CN202111251467.XA CN202111251467A CN113901726A CN 113901726 A CN113901726 A CN 113901726A CN 202111251467 A CN202111251467 A CN 202111251467A CN 113901726 A CN113901726 A CN 113901726A
- Authority
- CN
- China
- Prior art keywords
- seq2point
- model
- data
- training
- power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 18
- 238000012952 Resampling Methods 0.000 claims abstract description 12
- 238000005070 sampling Methods 0.000 claims abstract description 11
- 230000002159 abnormal effect Effects 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 32
- 210000002569 neuron Anatomy 0.000 claims description 25
- 238000012360 testing method Methods 0.000 claims description 9
- 238000005406 washing Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 3
- 230000002457 bidirectional effect Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000005315 distribution function Methods 0.000 claims description 2
- 238000011478 gradient descent method Methods 0.000 claims description 2
- 210000002364 input neuron Anatomy 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 claims description 2
- 210000004205 output neuron Anatomy 0.000 claims description 2
- 230000000717 retained effect Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- HTIQEAQVCYTUBX-UHFFFAOYSA-N amlodipine Chemical compound CCOC(=O)C1=C(COCCN)NC(C)=C(C(=O)OC)C1C1=CC=CC=C1Cl HTIQEAQVCYTUBX-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000004851 dishwashing Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/06—Power analysis or power optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a non-intrusive load decomposition method based on a seq2point model, which comprises S1, obtaining the total power of an ammeter corresponding to a target electrical equipment to be detected, and resampling the total power by adopting a preset sampling frequency to obtain resample data; s2, comparing the resampled data with the standard data of the total power of the electric meter, and deleting abnormal segments in the resampled data to obtain input data of the network; s3, calculating the optimal sequence length of the input data according to the initial length of the input data; s4, recognizing the input data adjusted to the optimal sequence length by adopting the trained seq2point model to obtain the power curve of the target electrical equipment to be measured.
Description
Technical Field
The invention relates to a power data decomposition technology, in particular to a non-intrusive load decomposition method based on a seq2point model.
Background
In recent years, smart meters are widely deployed around the world, massive fine-grained power consumption data are collected, stored and managed, a big data condition is provided for deep learning, an end-to-end processing scheme is provided for data processing, and for a non-invasive load monitoring task, load decomposition can be directly carried out without detecting an electric appliance switch event.
The existing seq2point model (Sequence-to-point model, see paper c.zhang, m.zhong, z.wang, n.goddard, and c.sutton, "Sequence-to-point learning with neural networks for non-intrusive load monitoring," in proc.32nd AAAI conf.artist. However, the identification and prediction performances of the electric appliance in a long-time multi-state are poor, and the identification and prediction performances of other equipment are still to be improved, namely, the overfitting problem exists, and the improvement of the network structure and the elimination of the overfitting are very important.
Disclosure of Invention
Aiming at the defects in the prior art, the non-intrusive load decomposition method based on the seq2point model solves the problem that the existing seq2point model is poor in prediction performance.
In order to achieve the purpose of the invention, the invention adopts the technical scheme that:
a non-invasive load decomposition method based on a seq2point model is provided, which comprises the following steps:
s1, acquiring the total power of an ammeter corresponding to the target electrical equipment to be detected, and resampling the total power by adopting a preset sampling frequency to obtain resample data;
s2, comparing the resampled data with the standard data of the total power of the electric meter, and deleting abnormal segments in the resampled data to obtain input data of the network;
s3, calculating the optimal sequence length of the input data according to the initial length of the input data:
ESL＝SL×SI×RF
wherein SL is the initial length of input data; SI is a sampling interval; RF is the resampling ratio; ESL is the optimal sequence length;
s4, recognizing the input data adjusted to the optimal sequence length by adopting the trained seq2point model to obtain a power curve of the target electrical equipment to be measured:
xτ＝Fp(Yt：t+W-1)+ε
wherein x isτThe midpoint of the current equipment output window; y ist：t+W-1A sliding window segment for input data; xt：t+W-1A sliding window sequence of target electrical equipment to be detected;
the network structure of the seq2point model comprises five convolutional layers which are connected in sequence, the output of the last convolutional layer is connected with the full-connection layer and the output layer in sequence, and a plugging layer is inserted between the adjacent convolutional layers and between the last convolutional layer and the full-connection layer.
The invention has the beneficial effects that:
1. the optimal sequence is selected, the optimal sequence length of each device can be obtained according to the power characteristics of each target electrical device by combining the sampling rate and the resampling rate, the sequence length of an input window of the network model is corrected, and the self-adaptive adjustment of the input of the network model is completed, so that the network model can learn the complete and accurate characteristics of the target electrical device, and the subsequent load decomposition is fully prepared.
2. Eliminating overfitting: the dropout layer is added after the convolution layer, so that the characteristic quantity of the middle layer can be reduced, redundancy is reduced, orthogonality among layers is better fitted, the identification accuracy of the target electrical equipment is further improved, and the accuracy of predicting the power curve change of the target electrical equipment is over 85%.
3. By adopting the scheme to carry out load decomposition, the identification and prediction performance of long-time multi-state equipment is improved, the close fit is carried out on the power curve, and the defect that the existing seq2point model cannot identify and predict the continuously variable equipment with power is overcome.
Drawings
Fig. 1 is a flowchart of a non-intrusive load decomposition method based on a seq2point model.
Fig. 2 is a comparison diagram of the predicted power curve and the actual power curve of the refrigerators in houses 1, 2, and 3 in the embodiment, (a) a comparison diagram of house 1, (b) a comparison diagram of house 2, and (c) a comparison diagram of house 3.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate the understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and it will be apparent to those skilled in the art that various changes may be made without departing from the spirit and scope of the invention as defined and defined in the appended claims, and all matters produced by the invention using the inventive concept are protected.
Referring to fig. 1, fig. 1 shows a flow chart of a non-intrusive load decomposition method based on a seq2point model; as shown in fig. 1, the method includes steps S1 through S4.
In step S1, obtaining total power of the electric meter corresponding to the target electric device to be tested, and resampling the total power by using a preset sampling frequency to obtain resample data; according to the scheme, the preset sampling frequency is preferably 8 s/time, and uniformity of granularity can be realized after resampling, so that the data wave bands of different electric meters and the data of the same target electric equipment under different electric meters are consistent.
The target electrical equipment comprises a kettle, a microwave oven, a refrigerator, a dish washing machine and a washing machine, and is of the same type when training.
In step S2, comparing the resampled data with the standard data of the total power of the electric meter, and deleting abnormal segments in the resampled data to obtain input data of the network;
in step S3, the optimal sequence length of the input data is calculated from the initial length of the input data:
ESL＝SL×SI×RF
wherein SL is the initial length of input data; SI is a sampling interval; RF is the resampling ratio; ESL is the optimal sequence length;
the load decomposition method of the scheme also comprises the step of filling a preset number of zeros at the head and the tail of the input data before inputting the input data with the optimal sequence length into the seq2point model.
In step S4, recognizing the input data adjusted to the optimal sequence length by using the trained seq2point model to obtain a power curve of the target electrical equipment to be measured:
xτ＝Fp(Yt：t+W-1)+ε
wherein x isτThe midpoint of the current equipment output window; y ist：t+W-1A sliding window segment for input data; xt：t+W-1A sliding window sequence of target electrical equipment to be detected;
the network structure of the seq2point model comprises five convolutional layers which are connected in sequence, the output of the last convolutional layer is connected with a full-connection layer and an output layer in sequence, and a plugging layer is inserted between the adjacent convolutional layers and between the last convolutional layer and the full-connection layer.
According to the scheme, the size of the input window of each device is adjusted by using the optimal sequence, so that the power section of each device can be input completely, and the complete learning of the characteristics of the target electrical equipment by the neural network is facilitated.
Meanwhile, the network construction of the original seq2point model is changed by using the dropout layer, so that the neurons in the network layer can be forced to abandon the connection of a specific neuron and search the connection with more other neurons, more common features are learned or the weight of some features is reduced, and the robustness of the network is improved; and the joint adaptability among all network layers is enabled, and the generalization capability is enhanced.
In addition, the input of the seq2point model of the scheme is a sliding window segment of the total load (total power), the output is a single point of the target electrical equipment, the representation capability of the network is concentrated in the middle point of the window, and more accurate prediction is generated.
In implementation, the loss function of the seq2point model is preferably as follows:
wherein, thetapNetwork parameters of the seq2point model; t is the sequence length; w is a window of the sliding window; log p (.) is the logarithm of the extracted features.
In an embodiment of the present invention, the method for training the seq2point model includes:
s41, acquiring a data set formed by total power of a plurality of electric meters connected with target electric equipment and power data of the corresponding target electric equipment;
the data set used in the scheme is REDD, UK-Dale, and the specific information is as follows:
UK-Dale dataset: the data set contains measurements of more than 10 types of appliances, which are only of interest for kettles, microwave ovens, refrigerators, dishwashers and washing machines, which are popular appliances for assessing non-intrusive power load monitoring methods, and was implemented using room 1, 3, 4, 5 training neural networks, room 2 as test data, since only room 1 and room 2 had these devices.
REDD dataset: the device and power readings are recorded every 3 seconds and 1 second, respectively. This data set contains the measured values from six families, and this scheme has used 2nd to 6 th house to train, and 1 # house carries out the algorithm test, owing to do not have kettle data, has only utilized microwave oven, refrigerator, dish washer and washing machine.
When the load decomposition is realized, if the power data of the refrigerator in the house needs to be monitored, then when the seq2point model is trained, the data set only needs the power data of the refrigerator and the total power of the corresponding ammeter, if the power data of the refrigerator and the washing machine of two target electrical equipment needs to be monitored, then a seq2point model for identifying the refrigerator and a seq2point model for identifying the washing machine need to be trained respectively during the training.
The typical devices used for training and testing in the scheme are five types in total, and the device names and the relevant important attribute values are listed as follows:
microwave: the power threshold is 200, the maximum power is 3969, the power mean is 500, and the power variance is 800;
fridge: the power threshold value is 50, the maximum power is 3323, the power mean value is 200, and the power variance is 400;
dishwasher: the power threshold is 10, the maximum power is 3964, the power mean is 700, and the power variance is 1000;
washingmachine: the power threshold is 20, the maximum power is 3999, the power mean is 400, and the power variance is 700;
and (5) a button: the power threshold is 2000, the maximum power is 3998, the power mean is 700, and the power variance is 1000.
S42, resampling the data in the data set by adopting a preset sampling frequency, comparing the resampled data with standard data, and deleting abnormal segments in the resampled data to obtain an updated data set;
s43, respectively calculating the optimal sequence length of each data according to the initial length of each data in the updated data set, and obtaining a training set with the optimal sequence length of each data;
s44, training the seq2point model by adopting a training set, and then testing the trained seq2point model by adopting a test set formed by power data in electric meters accessed with target electric equipment to obtain a power curve of each target electric equipment;
in implementation, the method preferably adopts a training set to train the seq2point model, and further comprises:
s441, randomly deleting half of hidden neurons in the seq2point model, and keeping the input and output neurons unchanged;
s442, inputting data in a preset proportion in the training set into a network with half of hidden neurons deleted for forward propagation, and performing backward propagation on the obtained loss result; updating network parameters on the retained neurons by adopting a random gradient descent method;
s443, recovering the deleted hidden neurons, and judging whether a network parameter loss function is smaller than a preset error or whether the iteration times are larger than a preset iteration times; if any condition is satisfied, the training of the seq2point model is completed, otherwise, the step S441 is returned.
Wherein, the calculation formula of the network in the training process of the seq2point model is as follows:
wherein r isj (l)Is a Bernoulli distribution value constructed according to the probability p; bernoulli (·) is a Bernoulli distribution function; p is the probability;
output for neurons that were discarded randomly; y is(l)Is the primary output of the neuron;
the output of the ith neuron at layer l + 1;
is the ith neuron in convolutional layerWeight at layer l + 1;
bias at layer l +1 for the ith neuron;
is the output of the neuron; f (.) is the activation function of the neuron.
S45, comparing the power curve of each target electrical equipment with the original power curve of the target electrical equipment, and obtaining the identification accuracy of the seq2point model according to the comparison result;
s46, judging whether the recognition accuracy is greater than the preset accuracy, if so, finishing the training of the seq2point model, otherwise, entering the step S47;
and S47, correcting the trained seq2point model by adopting a hole residual error deep layer network of a bidirectional attention mechanism to obtain the trained seq2point model.
Examples
Taking REDD data set as an example, identifying the device fridge, wherein the relevant attributes and data set of the fridge are divided as follows:
'fridge': { 'input window size': 599, 'power threshold': 50, 'maximum power': 3323 'power mean': 200, 'power variance': 400, 'sequence length': 512, 'training house serial number': [1, 2, 3], 'Power channel': [5,9,7]}
Extracting data of three channels, namely a channel 5 of the house 1, a channel 9 of the house 2 and a channel 7 of the house 3 in the Red, unifying granularity according to the resampling rate of 8 s/time, and solving for the optimal sequence length (ESL).
The extracted data is divided into a training set, a verification set and three test sets.
By adopting the scheme, the seq2point model is constructed, and the parameters in the model are set as follows: the training batch size is 1000, the training sample size is 100000, the number of training rounds is 5, the input window size is 599, the verification frequency is 1, the output window size is 1, the loss function is 'mean square error', the index is 'mse, mae', and the learning rate is 0.001.
After the relevant data set and model parameters are completed, the training set and the verification set are simultaneously input into the network model to start training, and relevant models and parameter information of the fridge are generated according to the training method of the scheme, wherein the relevant models and parameter information comprise, for example, Mean Square Error (MSE) of 0.0926, Mean Aggregate Error (MAE) of 0.2754, weight of the convolutional layer of 37200 and weight of the fully-connected layer of 30669824.
The mean square error (MAE) is calculated as:
wherein x istIn order to predict the value of the target,
are true values.
The Mean Square Error (MSE) is calculated as:
wherein r represents the true value of the compound,
which is representative of the inferred value(s),
in the testing process, a trained seq2point model about fridge is called, the total power of house 1, house 2 and house 3 is input, the total power is decomposed, a predicted power curve of the fridge in the house is obtained, the predicted power curve is compared with a real power curve, and the comparison result is shown in fig. 2.
It can be seen from the comparison diagram of fig. 2 that the kind of the device can be better identified by using the seq2point model of the optimal sequence, the change of the power curve can be better predicted, and the accuracy can reach more than 85% except for the power section of the device power jump.
Claims (7)
1. The non-invasive load decomposition method based on the seq2point model is characterized by comprising the following steps:
s1, acquiring the total power of an ammeter corresponding to the target electrical equipment to be detected, and resampling the total power by adopting a preset sampling frequency to obtain resample data;
s2, comparing the resampled data with the standard data of the total power of the electric meter, and deleting abnormal segments in the resampled data to obtain input data of the network;
s3, calculating the optimal sequence length of the input data according to the initial length of the input data:
ESL＝SL×SI×RF
wherein SL is the initial length of input data; SI is a sampling interval; RF is the resampling ratio; ESL is the optimal sequence length;
s4, recognizing the input data adjusted to the optimal sequence length by adopting the trained seq2point model to obtain a power curve of the target electrical equipment to be measured:
xτ＝Fp(Yt：t+W-1)+ε
wherein x isτThe midpoint of the current equipment output window; y ist：t+W-1A sliding window segment for input data; xt：t+W-1A sliding window sequence of the target electrical equipment;
the network structure of the seq2point model comprises five convolutional layers which are connected in sequence, the output of the last convolutional layer is connected with the full-connection layer and the output layer in sequence, and a plugging layer is inserted between the adjacent convolutional layers and between the last convolutional layer and the full-connection layer.
2. The method for non-intrusive load decomposition based on a seq2point model according to claim 1, wherein the loss function of the seq2point model is as follows:
wherein, thetapNetwork parameters of the seq2point model; t is the sequence length; w is a window of the sliding window; log p (.) is the logarithm of the extracted features.
3. The method for non-intrusive load decomposition based on the seq2point model according to claim 1 or 2, wherein the training method of the seq2point model comprises:
s41, acquiring a data set formed by total power of a plurality of electric meters connected with target electric equipment and power data of the corresponding target electric equipment;
s42, resampling the data in the data set by adopting a preset sampling frequency, comparing the resampled data with standard data, and deleting abnormal segments in the resampled data to obtain an updated data set;
s43, respectively calculating the optimal sequence length of each data according to the initial length of each data in the updated data set, and obtaining a training set with the optimal sequence length of each data;
s44, training the seq2point model by adopting a training set, and then testing the trained seq2point model by adopting a test set formed by power data in electric meters accessed with target electric equipment to obtain a power curve of each target electric equipment;
s45, comparing the power curve of each target electrical equipment with the original power curve of the target electrical equipment, and obtaining the identification accuracy of the seq2point model according to the comparison result;
s46, judging whether the recognition accuracy is greater than the preset accuracy, if so, finishing the training of the seq2point model, otherwise, entering the step S47;
and S47, correcting the trained seq2point model by adopting a hole residual error deep layer network of a bidirectional attention mechanism to obtain the trained seq2point model.
4. The method of claim 3, wherein the training of the seq2point model with the training set further comprises:
s441, randomly deleting half of hidden neurons in the seq2point model, and keeping the input and output neurons unchanged;
s442, inputting data in a preset proportion in the training set into a network with half of hidden neurons deleted for forward propagation, performing backward propagation on the obtained loss result, and then updating network parameters on the retained neurons by adopting a random gradient descent method;
s443, recovering the deleted hidden neurons, and judging whether a network parameter loss function is smaller than a preset error or whether the iteration times are larger than a preset iteration times; if any condition is satisfied, the training of the seq2point model is completed, otherwise, the step S441 is returned.
5. The method of claim 4, wherein the calculation formula of the network in the training process of the seq2point model is as follows:
wherein r isj (l)Is a Bernoulli distribution value constructed according to the probability p; bernoulli (·) is a Bernoulli distribution function; p is the probability;
output for neurons that were discarded randomly; y is(l)Is the primary output of the neuron;
the output of the ith neuron at layer l + 1;
the weight of the ith neuron in the convolutional layer at the l +1 layer;
bias at layer l +1 for the ith neuron;
is the output of the current neuron; f (.) is the output function of the neuron.
6. The method of claim 1, 2, 4 or 5, wherein the step of inputting the input data with the optimal sequence length into the seq2point model further comprises filling a preset number of zeros at the head and tail of the input data.
7. The seq2point model-based non-intrusive load decomposition method according to claim 1, 2, 4 or 5, wherein the target electrical appliances comprise a kettle, a microwave oven, a refrigerator, a dishwasher and a washing machine, and the target electrical appliances are the same type of appliances when training is performed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111251467.XA CN113901726B (en) | 2021-10-27 | 2021-10-27 | Non-invasive load decomposition method based on seq2point model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111251467.XA CN113901726B (en) | 2021-10-27 | 2021-10-27 | Non-invasive load decomposition method based on seq2point model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113901726A true CN113901726A (en) | 2022-01-07 |
| CN113901726B CN113901726B (en) | 2024-06-21 |
Family
ID=79026886
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111251467.XA Active CN113901726B (en) | 2021-10-27 | 2021-10-27 | Non-invasive load decomposition method based on seq2point model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113901726B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330553A (en) * | 2022-06-22 | 2022-11-11 | 四川大学 | Equipment feature based multi-layer optimal selection non-intrusive load decomposition method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852655A (en) * | 2019-11-29 | 2020-02-28 | 湖南科技大学 | Household power load decomposition system with solar power supply system and decomposition method |
| CN111199014A (en) * | 2019-12-27 | 2020-05-26 | 国家电网有限公司 | Time sequence based seq2point NILM method and device |
| CN111753968A (en) * | 2020-07-07 | 2020-10-09 | 苏州联电能源发展有限公司 | Non-invasive load monitoring intelligent ammeter and electricity decomposition method |
| CN112215406A (en) * | 2020-09-23 | 2021-01-12 | 国网甘肃省电力公司营销服务中心 | Non-invasive type residential electricity load decomposition method based on time convolution neural network |
| US10983496B1 (en) * | 2020-10-13 | 2021-04-20 | King Abdulaziz University | Apparatus and method for non-intrusive load monitoring aided hybrid high and low frequency approaches |
-
2021
- 2021-10-27 CN CN202111251467.XA patent/CN113901726B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852655A (en) * | 2019-11-29 | 2020-02-28 | 湖南科技大学 | Household power load decomposition system with solar power supply system and decomposition method |
| CN111199014A (en) * | 2019-12-27 | 2020-05-26 | 国家电网有限公司 | Time sequence based seq2point NILM method and device |
| CN111753968A (en) * | 2020-07-07 | 2020-10-09 | 苏州联电能源发展有限公司 | Non-invasive load monitoring intelligent ammeter and electricity decomposition method |
| CN112215406A (en) * | 2020-09-23 | 2021-01-12 | 国网甘肃省电力公司营销服务中心 | Non-invasive type residential electricity load decomposition method based on time convolution neural network |
| US10983496B1 (en) * | 2020-10-13 | 2021-04-20 | King Abdulaziz University | Apparatus and method for non-intrusive load monitoring aided hybrid high and low frequency approaches |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330553A (en) * | 2022-06-22 | 2022-11-11 | 四川大学 | Equipment feature based multi-layer optimal selection non-intrusive load decomposition method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113901726B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111695620B (en) | Method and system for detecting and correcting time sequence abnormal data of power system | |
| CN111639586B (en) | Non-invasive load identification model construction method, load identification method and system | |
| CN112787736B (en) | Long-short term memory cooperative spectrum sensing method based on covariance matrix | |
| CN111339129A (en) | Remote meter reading abnormity monitoring method and device, gas meter system and gas cloud platform | |
| CN113901726B (en) | Non-invasive load decomposition method based on seq2point model | |
| CN111783684A (en) | Bi-GRU-SA-based household electrical load decomposition method and system | |
| CN108924847B (en) | Cognitive radio frequency spectrum prediction method and device based on ANN | |
| CN109542952A (en) | A kind of detection method of time series abnormal point | |
| CN111898637B (en) | Feature selection algorithm based on Relieff-DDC | |
| CN108959188A (en) | Granger Causality based on quantization minimal error entropy criterion sentences the method for distinguishing | |
| CN113484813A (en) | Intelligent ammeter fault rate estimation method and system under multi-environment stress | |
| CN117076955A (en) | Fault detection method and system for high-voltage frequency converter | |
| CN112671585A (en) | Intelligent household equipment exception handling method and device, processor and electronic equipment | |
| CN114124734A (en) | Network traffic prediction method based on GCN-Transformer integration model | |
| CN116361709B (en) | Self-adaptive power load identification method and device | |
| CN113098640B (en) | Frequency spectrum anomaly detection method based on channel occupancy prediction | |
| CN116167008A (en) | Abnormal positioning method for internet of things sensing cloud data center based on data enhancement | |
| CN115330553A (en) | Equipment feature based multi-layer optimal selection non-intrusive load decomposition method | |
| CN115422603A (en) | Block cipher hardware security assessment method based on template principal component regression | |
| CN113642421A (en) | Event detection method, system, equipment and storage medium for periodic multi-state electric appliance | |
| CN112465150A (en) | Real data enhancement-based multi-element time sequence data filling method | |
| Levasseur et al. | Time Series Representation for Real-World Applications of Deep Neural Networks | |
| CN117295149B (en) | Frame synchronization method and system based on low-complexity ELM assistance | |
| CN112766459B (en) | Abnormality detection method based on generator | |
| Cao et al. | Graph Active Learning at Subgraph Granularity |
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 |