CN115828091A - Non-invasive load identification method and system based on end-cloud cooperation - Google Patents
Non-invasive load identification method and system based on end-cloud cooperation Download PDFInfo
- Publication number
- CN115828091A CN115828091A CN202211121833.4A CN202211121833A CN115828091A CN 115828091 A CN115828091 A CN 115828091A CN 202211121833 A CN202211121833 A CN 202211121833A CN 115828091 A CN115828091 A CN 115828091A
- Authority
- CN
- China
- Prior art keywords
- model
- days
- load
- data
- identification
- 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
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a non-invasive load identification method and a non-invasive load identification system based on end-cloud cooperation, wherein the method comprises the following steps: acquiring power consumption data from non-invasive load acquisition equipment, identifying the load type by using a first model on terminal equipment, and directly returning a result if the identification result is of a certain electrical appliance type; and if the identification result is other electrical appliances, the electricity utilization data is sent to the cloud platform, and the second model on the cloud platform is used for carrying out load identification to obtain the specific electrical appliance type. The method provided by the invention can effectively solve the problem of low identification efficiency of the terminal equipment caused by insufficient computing resources, and has good application value.
Description
Technical Field
The invention relates to identification of non-intrusive load data, in particular to a non-intrusive load identification method and system based on end-cloud cooperation.
Background
With the advance of the national "double carbon" policy and the rising demand of residential electricity, intelligent electricity management becomes an important component for enhancing the smart grid, wherein a non-intrusive load identification technology is one of the key technologies. Through installing non-invasive load collection equipment, can realize carrying out whole collection to power consumption information such as electric current, voltage, power. Through analyzing the power load data of different types, the identification of the electrical appliances of the residential users can be realized. Compared with an invasive technology, the non-invasive load identification technology has the remarkable advantages of being user-friendly, low in cost, convenient to install and maintain and the like, and has wide application prospects.
Currently, the identification of the non-intrusive loads is generally performed on the local terminal equipment. With the rapid increase of the number and types of residential electric devices, the demand for the efficiency of electric data processing and device monitoring is continuously increasing, and the computing resources of local devices are relatively limited. Therefore, the problems of long identification time and low identification efficiency generally exist. How to improve the recognition efficiency of the device becomes one of the problems to be solved currently.
Disclosure of Invention
The invention aims to solve the technical problems in the prior art, provides a non-invasive load identification method and system based on end-cloud cooperation, and improves the non-invasive identification efficiency.
In order to achieve the technical purpose, the invention adopts the following technical scheme.
In a first aspect, the invention discloses a non-invasive load identification method based on end-cloud cooperation, which comprises the following steps: acquiring power consumption data from non-invasive load acquisition equipment, identifying the load type by using a first model on terminal equipment, and directly returning a result if the identification result is of a certain electrical appliance type; and if the identification result is other electrical appliances, the electricity utilization data is sent to the cloud platform, and the second model on the cloud platform is used for carrying out load identification to obtain the specific electrical appliance type.
Further, the training method of the first model comprises the following steps:
acquiring resident electricity consumption data from non-invasive load equipment, marking the resident electricity consumption data, counting N electric appliances with the highest utilization rate, and setting the types of other electric appliances as other electric appliances; carrying out data preprocessing on the electricity consumption data of residents;
dividing the preprocessed data set, selecting data from 30 days to 10 days ago as a training set, selecting data from 10 days to 5 days ago as a testing set, and selecting data from 5 days ago as a verification set;
respectively extracting the maximum value, the minimum value, the average value and the median value of the current, the voltage and the active power corresponding to each record in the training set, the test set and the verification set in the same time interval of the previous 5 days, the previous 15 days and the previous 30 days as characteristics, and using the characteristics as the input of a non-invasive load model;
calculating the proportion E of the number of samples of each electrical appliance type to the number of samples of the whole training set i The weight value is used as the weight value of the loss function of the next iteration;
and performing cyclic iteration training on the non-invasive load model, and when the iteration times reach a set value or the error output by the non-invasive load model meets the requirement, obtaining a first model meeting the requirement.
Still further, the loss function is expressed as follows:
wherein E i The number of samples representing the electrical type to which the ith sample belongs accounts for the total number of samples in the whole training set, m is the total number of samples in the training set,
is the output value of the non-invasive load model, y i Are true values.
Further, the training method of the second model comprises:
acquiring resident electricity consumption data from non-invasive load equipment, and performing data preprocessing on the resident electricity consumption data;
dividing the preprocessed data set, selecting data from 30 days to 10 days ago as a training set, selecting data from 10 days to 5 days ago as a testing set, and selecting data from 5 days ago as a verification set;
respectively extracting the maximum value, the minimum value, the average value and the median value of the current, the voltage and the active power corresponding to each record in the training set, the testing set and the verifying set in the same time interval of the previous 5 days, the previous 15 days and the previous 30 days as characteristics, and using the characteristics as the input of a plurality of non-invasive load models;
performing cyclic iteration training on a plurality of non-invasive load models by taking the root mean square error as an evaluation index, and obtaining a plurality of non-invasive load models meeting the requirement when the iteration times reach a set value or the error output by the non-invasive load models meets the requirement;
and verifying the plurality of non-invasive load models by using a verification set, and selecting the model with the minimum root mean square error as the second model.
Further, the non-invasive load identification model is one of a CO regression analysis model, a factor hidden Markov model, a long-short term memory model, a recurrent neural network GRU, a Seq2Seq model, a Seq2point model and a BERT model.
In a second aspect, a non-intrusive load identification system based on end-cloud coordination includes: the system comprises a platform and terminal equipment, wherein a first model is arranged on the terminal equipment, the terminal equipment is used for collecting power utilization data from non-invasive load collection equipment, load type identification is realized by using the first model on the terminal equipment, and if an identification result is of a certain electrical appliance type, the result is directly returned; if the identification result is other electrical appliances, the electricity utilization data are sent to a cloud platform;
the platform is used for carrying out load identification by utilizing the second model to obtain a specific electric appliance type. The invention has the following beneficial technical effects:
the method comprises the steps of firstly, identifying by using a simplified model, and directly returning a result when an identification result is within the range of N electric appliances; and if the identification result is 'other electrical appliances', sending the data into the cloud platform for further identification to obtain a specific electrical appliance type, so that non-invasive load identification of end-cloud cooperation is realized. The method provided by the invention can effectively solve the problem of low identification efficiency of the terminal equipment caused by insufficient computing resources, and has good application value.
Drawings
Fig. 1 is a schematic flow chart of a non-intrusive load identification method based on end-cloud coordination according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, the present invention relates to a non-invasive load identification method and system based on end-cloud cooperation, which includes two parts, namely a terminal and a cloud platform, and mainly includes the following steps:
step 1, collecting power consumption data of not less than 50 residential users in nearly 30 days from non-invasive load equipment, and manually marking electric appliance labels;
step 1-1, the frequency of the collected data of the non-invasive load equipment is 1Hz;
step 1-2, the resident electricity consumption data comprise electricity consumption information such as current, voltage, active power and the like;
and 1-3, manually marking the collected data with an electric appliance label.
Step 2, preprocessing and characteristic extraction are carried out on the data obtained in the step 1, the data are used as input of a load identification algorithm, a complete model of non-invasive load identification is constructed, and the model is deployed on a cloud platform;
step 2-1, preprocessing the data collected in the step 1, performing noise reduction processing on the data to avoid the influence of noise data on model training, and then performing normalization operation on the data to improve the training effect of the model;
step 2-2, segmenting the training set, the test set and the verification set, selecting data from 30 days to 10 days ago as the training set, selecting data from 10 days to 5 days ago as the test set, and selecting data from 5 days ago to the present as the verification set;
2-3, respectively extracting the maximum value, the minimum value, the average value and the median value of the current, the voltage and the active power in each record in the training set, the test set and the verification set in the same time interval of the previous 5 days, the previous 15 days and the previous 30 days as characteristics, and using the characteristics as the input of a non-invasive load algorithm;
step 2-4, training load identification models by using the training set characteristics extracted in the step 2-3 as input, using Root Mean Square Error (RMSE) as evaluation indexes and using different parameters respectively, wherein the used non-invasive load identification algorithm is one of CO, FHMM, LSTM, GRU, seq2Seq, seq2point and BERT;
step 2-5, evaluating the effect of the different models trained in the step 2-4 on the verification set, selecting the model with the minimum RMSE as the optimal model, and taking the corresponding parameters as the optimal parameters;
step 2-6, taking the optimal model selected in the step 2-5 as a complete model for non-invasive load identification, and deploying the optimal model on a cloud platform;
step 3, counting N electric appliances with the highest utilization rate, setting the types of other electric appliances as other electric appliances, preprocessing and extracting characteristics of the data, constructing a simplified model for non-invasive load identification by using a load identification algorithm, and deploying the simplified model to terminal equipment;
step 3-1, counting the N electric appliances with the highest utilization rate in the resident electricity consumption data collected in the step 1;
step 3-2, preprocessing the data obtained by statistics in the step 3-1, denoising the data, avoiding the influence of noise data on model training, and then normalizing the data to improve the training effect of the model;
3-3, dividing the data set processed in the step 3-2, selecting data from 30 days to 10 days ago as a training set, selecting data from 10 days to 5 days ago as a testing set, and selecting data from 5 days ago as a verification set;
3-4, respectively extracting the maximum value, the minimum value, the average value and the median value of the current, the voltage and the active power corresponding to each record in the training set, the test set and the verification set in the same time interval of the previous 5 days, the previous 15 days and the previous 30 days as characteristics, and using the characteristics as the input of a non-invasive load algorithm;
3-5, training a load identification model by using the training set characteristics extracted in the step 3-4 as input and Root Mean Square Error (RMSE) as an evaluation index and using different parameters respectively, wherein the used non-invasive load identification algorithm is one of CO, FHMM, LSTM, GRU, seq2Seq, seq2point and BERT;
step 3-6, evaluating the effect of the different models trained in the step 3-5 on the verification set, selecting the model with the minimum RMSE as the optimal model, and taking the corresponding parameters as the optimal parameters;
3-7, taking the optimal model selected in the step 3-6 as a simplified model of a non-invasive load, and deploying the simplified model to terminal equipment;
and 4, acquiring power utilization data from the non-invasive load acquisition equipment, identifying the load type by using a simplified model on the terminal equipment, and directly returning the result if the identification result is a specific electric appliance type. And if the identification result is 'other electrical appliances', the electricity utilization data is sent to the cloud platform, and the load identification is carried out by the complete model on the cloud platform to obtain the specific electrical appliance type.
The process of the present invention is illustrated below with reference to examples.
The non-intrusive load identification method based on end-cloud cooperation in the embodiment comprises the following steps:
step 1, collecting power consumption data ED1 of 50 residential users in a certain cell within 30 days by using non-invasive load equipment and taking 1Hz as a sampling frequency, wherein the power consumption information comprises current, voltage and active power, and then manually marking an electric appliance label;
step 2, preprocessing and feature extraction are carried out on the ED1, the processed ED1 is used as the input of a load identification algorithm, a complete model (namely a second model) of non-intrusive load identification is constructed and deployed on a cloud platform:
step 2-1, preprocessing ED1, firstly carrying out noise reduction processing on data, then carrying out normalization operation on the data, and finally obtaining processed data DS1;
2-2, selecting data from 30 days to 10 days before in the data set DS1 as a training set, selecting data from 10 days to 5 days before as a test set, and selecting data from 5 days before as a verification set;
2-3, respectively extracting the maximum value, the minimum value, the average value and the median value of the same time interval of the first 5 days, the first 15 days and the first 30 days of the current, the voltage and the active power corresponding to each record in the training set, the test set and the verification set as features to form a feature set FS1;
step 2-4, the obtained feature set FS1 is used as input, root Mean Square Error (RMSE) is used as an evaluation index, different parameters are respectively used for training a load identification model, and the used non-intrusive load identification algorithm is one of CO, FHMM, LSTM, GRU, seq2Seq, seq2point and BERT;
2-5, selecting the model with the minimum RMSE as an optimal model M1 and the corresponding parameters as optimal parameters by using the scores of the non-invasive load identification models with different parameters trained in the step 2-4 on a verification set;
step 2-6, taking the M1 as a complete model of non-invasive load identification, and deploying the model on a cloud platform;
step 3, counting the 5 electrical appliances with the highest utilization rate in the data set ED1, setting the types of the other electrical appliances as other electrical appliances, preprocessing and extracting characteristics of the data, constructing a simplified model for non-invasive load identification by using a load identification algorithm, and deploying the simplified model to terminal equipment:
step 3-1, analyzing ED1, and counting to obtain 5 electric appliances with the highest utilization rate: refrigerators, microwave ovens, televisions, water heaters, washing machines;
step 3-2, extracting the data of the 5 electrical appliances in the step 3-1 from the ED1, and performing noise reduction and normalization operation to obtain a processed data set DS2;
3-3, selecting data from 30 days to 10 days before in the data set DS2 as a training set, selecting data from 10 days to 5 days before as a test set, and selecting data from 5 days before as a verification set;
3-4, respectively extracting the maximum value, the minimum value, the average value and the median value of the same time interval of the first 5 days, the first 15 days and the first 30 days of the current, the voltage and the active power corresponding to each record in the training set, the test set and the verification set as features to form a feature set FS2;
step 2-4, using the obtained feature set FS2 as input, using Root Mean Square Error (RMSE) as an evaluation index, and respectively using different parameters to train a load identification model, wherein the used non-intrusive load identification algorithm is one of CO, FHMM, LSTM, GRU, seq2Seq, seq2point and BERT;
3-5, selecting the model with the minimum RMSE as an optimal model M2 and the corresponding parameters as the optimal parameters by using the scores of the non-invasive load identification models with different parameters trained in the step 2-4 on the verification set;
3-6, taking the M2 as a simplified model (namely a first model) of non-invasive load identification, and deploying the simplified model to the terminal equipment;
and 4, acquiring power consumption data ED2 from non-invasive load acquisition equipment, performing noise reduction and normalization pretreatment to obtain a data set DS3, identifying the load type by using a simplified model M2 on terminal equipment, and directly returning a result if the identification result is one of electric appliances such as a refrigerator, a microwave oven, a television, a water heater and a washing machine. And if the identification result is 'other electrical appliances', the electricity utilization data DS3 is sent to the cloud platform, and the complete model M1 on the cloud platform carries out load identification to obtain the specific electrical appliance type and return the result.
The invention discloses a non-invasive load identification method and a non-invasive load identification system based on terminal-cloud cooperation, which are characterized in that firstly, user electricity consumption data are obtained from non-invasive load acquisition equipment, and preprocessing and manual labeling of an electric appliance label are carried out; secondly, after the data set is divided, the electricity utilization characteristics of different electric appliance types are described according to the statistical characteristics of the maximum value, the minimum value, the average value, the median value and the like of the same time interval of the current, the voltage and the active power corresponding to each record in the previous 5 days, the previous 15 days and the previous 30 days, so that a complete model of load identification is trained according to the characteristics and is deployed to a cloud platform; and then counting the N electric appliances with the highest utilization rate, training the simplified model of load identification by using the data of the electric appliances, and deploying the simplified model to the terminal equipment. In the process of identifying a new electricity utilization data set, firstly, a simplified model is used for identification, and when the identification result is within the range of N electrical appliances, the result is directly returned; and if the identification result is 'other electrical appliances', sending the data to a cloud platform for further identification to obtain a specific electrical appliance type, thereby realizing non-invasive load identification of end-cloud cooperation. The method of the invention can effectively solve the problem of low identification efficiency of the terminal equipment caused by insufficient computing resources, and has good application value.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the system/module described above may refer to the corresponding process in the foregoing method embodiment, and is not described herein again.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and those skilled in the art can make various modifications without departing from the spirit and scope of the present invention as defined in the appended claims.
Claims (6)
1. The non-intrusive load identification method based on end-cloud cooperation is characterized by comprising the following steps:
acquiring power consumption data from non-invasive load acquisition equipment, identifying the load type by using a first model on terminal equipment, and directly returning a result if the identification result is of a certain electrical appliance type; and if the identification result is other electrical appliances, the electricity utilization data is sent to the cloud platform, and the second model on the cloud platform is used for carrying out load identification to obtain the specific electrical appliance type.
2. The method for non-intrusive load identification based on end-cloud coordination according to claim 1, wherein the training method of the first model comprises:
acquiring resident electricity consumption data from non-invasive load equipment, marking the resident electricity consumption data, counting N electric appliances with the highest utilization rate, and setting the types of other electric appliances as other electric appliances; carrying out data preprocessing on the electricity consumption data of residents;
dividing the preprocessed data set, selecting data from 30 days to 10 days ago as a training set, selecting data from 10 days to 5 days ago as a testing set, and selecting data from 5 days ago as a verification set;
respectively extracting the maximum value, the minimum value, the average value and the median value of the current, the voltage and the active power corresponding to each record in the training set, the test set and the verification set in the same time interval of the previous 5 days, the previous 15 days and the previous 30 days as characteristics, and using the characteristics as the input of a non-invasive load model;
calculating the proportion E of the number of samples of each electrical appliance type to the number of samples of the whole training set i The weight value is used as the weight value of the loss function of the next iteration;
and performing loop iteration training on the non-intrusive load model, and obtaining a first model meeting the requirement when the iteration times reach a set value or the error output by the non-intrusive load model meets the requirement.
3. The method for non-intrusive load identification based on end-cloud coordination according to claim 2, wherein the expression of the loss function is as follows:
wherein E i The number of samples representing the electrical type to which the ith sample belongs accounts for the total number of samples in the entire training set, m is the total number of samples in the training set,
is the output value of the non-invasive load model, y i Are true values.
4. The method for non-intrusive load identification based on end-cloud coordination according to claim 1, wherein the training method of the second model comprises:
acquiring resident electricity consumption data from non-invasive load equipment, and performing data preprocessing on the resident electricity consumption data;
dividing the preprocessed data set, selecting data from 30 days to 10 days ago as a training set, selecting data from 10 days to 5 days ago as a testing set, and selecting data from 5 days ago as a verification set;
respectively extracting the maximum value, the minimum value, the average value and the median value of the current, the voltage and the active power corresponding to each record in the training set, the test set and the verification set in the same time interval of the previous 5 days, the previous 15 days and the previous 30 days as characteristics, and using the characteristics as the input of a plurality of non-invasive load models;
performing cyclic iteration training on a plurality of non-invasive load models by taking the root mean square error as an evaluation index, and obtaining a plurality of non-invasive load models meeting the requirement when the iteration times reach a set value or the error output by the non-invasive load models meets the requirement;
and verifying the plurality of non-invasive load models by using a verification set, and selecting the model with the minimum root mean square error as the second model.
5. The method for identifying the non-intrusive load based on the end-cloud coordination according to any one of claims 2 to 4, wherein the non-intrusive load identification model is one of a CO regression analysis model, a factor hidden Markov model, a long-short term memory model, a recurrent neural network GRU, a Seq2Seq model, a Seq2point model and a BERT model.
6. Non-intrusive load identification system based on end-cloud cooperation is characterized by comprising: the terminal equipment is used for collecting power utilization data from non-invasive load collection equipment, load type identification is realized by using the first model on the terminal equipment, and if the identification result is of a certain electrical appliance type, the result is directly returned; if the identification result is other electrical appliances, the electricity utilization data is sent to a cloud platform;
the platform is used for carrying out load identification by utilizing the second model to obtain a specific electric appliance type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211121833.4A CN115828091A (en) | 2022-09-15 | 2022-09-15 | Non-invasive load identification method and system based on end-cloud cooperation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211121833.4A CN115828091A (en) | 2022-09-15 | 2022-09-15 | Non-invasive load identification method and system based on end-cloud cooperation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115828091A true CN115828091A (en) | 2023-03-21 |
Family
ID=85523633
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211121833.4A Pending CN115828091A (en) | 2022-09-15 | 2022-09-15 | Non-invasive load identification method and system based on end-cloud cooperation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115828091A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117239743A (en) * | 2023-11-15 | 2023-12-15 | 青岛鼎信通讯股份有限公司 | Electric energy meter electricity load acquisition method, device, equipment and medium |
-
2022
- 2022-09-15 CN CN202211121833.4A patent/CN115828091A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117239743A (en) * | 2023-11-15 | 2023-12-15 | 青岛鼎信通讯股份有限公司 | Electric energy meter electricity load acquisition method, device, equipment and medium |
| CN117239743B (en) * | 2023-11-15 | 2024-02-27 | 青岛鼎信通讯股份有限公司 | Electric energy meter electricity load acquisition method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110956220B (en) | Non-invasive household appliance load identification method | |
| CN109598451B (en) | Non-invasive load identification method based on PCA (principal component analysis) and LSTM (least Square TM) neural network | |
| CN109492667A (en) | A kind of feature selecting discrimination method for non-intrusive electrical load monitoring | |
| CN110287762B (en) | Non-invasive load identification method and device based on data mining technology | |
| CN109840691B (en) | Non-invasive polynomial electric quantity estimation method based on deep neural network | |
| CN110991263B (en) | Non-invasive load identification method and system for resisting background load interference | |
| CN111489188B (en) | Resident adjustable load potential mining method and system | |
| CN110309884A (en) | Electricity consumption data anomalous identification system based on ubiquitous electric power Internet of Things net system | |
| CN110321934A (en) | Method and system for detecting abnormal data of user electricity consumption | |
| CN111242161A (en) | Non-invasive non-resident user load identification method based on intelligent learning | |
| CN111563827A (en) | Load decomposition method based on electrical appliance physical characteristics and residential electricity consumption behaviors | |
| CN115456034A (en) | Automatic identification and monitoring method and system for electric bicycle charging | |
| CN115828091A (en) | Non-invasive load identification method and system based on end-cloud cooperation | |
| CN113887912A (en) | Non-invasive load identification method for deeply learning downward embedded equipment | |
| CN113076985A (en) | Power consumer load identification method and device and computer readable medium | |
| CN117559443A (en) | Ordered power utilization control method for large industrial user cluster under peak load | |
| CN115758188A (en) | Non-invasive load identification method, device, equipment and medium | |
| CN114595762A (en) | Photovoltaic power station abnormal data sequence extraction method | |
| CN116995734B (en) | Distributed energy power quality monitoring control evaluation system for power grid | |
| CN105445577A (en) | Power quality disturbance source working condition identifying method | |
| CN111506636A (en) | System and method for analyzing residential electricity consumption behavior based on autoregressive and neighbor algorithm | |
| CN116502149A (en) | Low-voltage power distribution network user-transformation relation identification method and system based on current characteristic conduction | |
| CN111199014A (en) | Time sequence based seq2point NILM method and device | |
| CN107194529B (en) | Power distribution network reliability economic benefit analysis method and device based on mining technology | |
| CN115983347A (en) | Non-invasive load decomposition method, device and storage medium |
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 |