CN111327118B - Non-invasive power load identification method - Google Patents

Non-invasive power load identification method Download PDF

Info

Publication number
CN111327118B
CN111327118B CN202010243561.XA CN202010243561A CN111327118B CN 111327118 B CN111327118 B CN 111327118B CN 202010243561 A CN202010243561 A CN 202010243561A CN 111327118 B CN111327118 B CN 111327118B
Authority
CN
China
Prior art keywords
signal
sample
clustering
current
characteristic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010243561.XA
Other languages
Chinese (zh)
Other versions
CN111327118A (en
Inventor
梁炎明
刘倩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian University of Technology
Original Assignee
Xian University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xian University of Technology filed Critical Xian University of Technology
Priority to CN202010243561.XA priority Critical patent/CN111327118B/en
Publication of CN111327118A publication Critical patent/CN111327118A/en
Application granted granted Critical
Publication of CN111327118B publication Critical patent/CN111327118B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R21/00—Arrangements for measuring electric power or power factor
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R22/00—Arrangements for measuring time integral of electric power or current, e.g. electricity meters
    • G01R22/06—Arrangements for measuring time integral of electric power or current, e.g. electricity meters by electronic methods
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B70/00—Technologies for an efficient end-user side electric power management and consumption
    • Y02B70/30—Systems integrating technologies related to power network operation and communication or information technologies for improving the carbon footprint of the management of residential or tertiary loads, i.e. smart grids as climate change mitigation technology in the buildings sector, including also the last stages of power distribution and the control, monitoring or operating management systems at local level

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a non-invasive power load identification method, which comprises the following steps: step 1: collecting a current signal or a power meter signal of the electric appliance to obtain a current characteristic of the electric appliance, wherein the current characteristic is used as a sample characteristic in a characteristic library; step 2: decomposing the current characteristics of the electric appliance by using an empirical mode decomposition method to obtain sample characteristics to be classified; step 3: optimizing the characteristics of the sample to be classified by adopting a k-means clustering method; step 4: and identifying by using a Euclidean distance nearest matching principle. The method solves the problem that in the prior art, when noise signals exist, the power load identification accuracy is not high. The problem of when the active reactive power of the electrical apparatus is close, discern the effect relatively poor is solved. The accuracy and the decomposition efficiency of the non-invasive power load decomposition are improved.

Description

Non-invasive power load identification method
Technical Field
The invention belongs to the technical field of intelligent electricity utilization and energy efficiency monitoring, and particularly relates to a non-invasive power load identification method.
Background
The non-invasive power load identification technology is also called as a non-invasive power load monitoring technology, and is characterized in that a monitoring device is arranged at an entrance of a resident power user to collect power consumption information of a user main port, and the information of the total current and the total voltage at the entrance is analyzed to identify the start-stop state and the working state of each or each type of power consumption device of the resident user. By the technology, the use result of the electric appliance can be fed back to the user so as to guide the electricity consumption behavior of the user, thereby being beneficial to promoting the user to save electricity and realizing energy conservation and consumption reduction. Compared with the traditional invasive monitoring technology, the technology is realized without installing a large number of sensors or monitoring equipment, so that the installation cost and the time and the expense of later maintenance are greatly reduced. The device has the advantages of simple structure, convenient installation and debugging, low economic cost, wide coverage range and strong reliability. For individual users, the working state of the electric appliance can be known in time according to the monitoring information, so that the electric behavior can be optimized independently, and the electric energy can be saved; meanwhile, the device can help a user to quickly and accurately clear the faults of the electric appliance, and improves the life quality. For factories, public building sites: the planning scheme can be scientifically formulated and the electricity utilization proposal can be provided by the technology; accurately monitoring and positioning the power failure; and setting a safety capacity alarm, ensuring the safety of a power line and finding out hidden electricity consumption conditions. In the aspect of energy efficiency intelligent monitoring management: the technology plays an important role in load real-time electric quantity statistics, historical data query analysis, electricity consumption prediction, energy consumption evaluation, electricity consumption behavior monitoring, energy saving strategy execution condition checking, data-driven advanced analysis and the like. The technology has very important use value and research significance for green, continuous and coordinated development, energy production promotion and consumption revolution in China.
The existing non-invasive power load identification technology mainly establishes a power load model according to the electrical characteristics of the power load, then utilizes pattern identification and optimization technology to realize the decomposition of the power load, and most of identification is aimed at the identification of a single electric appliance. Meanwhile, most researches do not well solve the problem of load decomposition in the mixed operation of multiple types of loads under the interference of high-power non-stationary load fluctuation. In the actual use process, the recognition rate of the existing load of the interference signal is greatly reduced due to noise.
Disclosure of Invention
The invention aims to provide a non-invasive power load identification method, which solves the problem of low accuracy of power load identification when noise signals and interference signals exist in the prior art.
The technical scheme adopted by the invention is that the non-invasive power load identification method comprises the following steps:
step 1: collecting a current signal or a power meter signal of the electric appliance to obtain a current characteristic of the electric appliance, wherein the current characteristic is used as a sample characteristic in a characteristic library;
step 2: decomposing the current characteristics of the electric appliance by using an empirical mode decomposition method to obtain sample characteristics to be classified;
step 3: optimizing the characteristics of the sample to be classified by adopting a k-means clustering method;
step 4: and identifying by using a Euclidean distance nearest matching principle.
The invention is also characterized in that:
the specific process of the step 2 is as follows:
step 2.1: determining the length L of the decomposed signal
Defining a period of acquisition signals by at least 600 load data points, wherein the length of a signal to be decomposed is at least the signal length of one period, the length L of the signal is selected, and the current signal of an electric appliance with the length L is x (t);
step 2.2: drawing an envelope curve of the signal according to the maximum value and the minimum value point of the signal, calculating the average value of the upper envelope curve and the lower envelope curve of the signal, and obtaining a signal h 1 (t)；
Step 2.3: taking standard deviation of two successively screened decomposition signals as a judgment criterion of iteration stop, and taking h 1 (t) repeating step 2.2 as signal x (t) for k times until signal h is obtained 1k (t) stopping iteration if a criterion for stopping iteration is satisfied;
step 2.4: will signal h 1k (t) separating from the original signal, for the remaining signal component r 1 And (t) performing EMD decomposition to obtain the characteristics of the sample to be classified.
The specific process of the step 3 is as follows:
step 3.1: determining the number K of the categories of the clusters; k=2 n ；
Step 3.2: and arbitrarily selecting a group of load characteristics from each decomposed load sample as an initial center of the cluster.
The specific process of the step 4 is as follows:
step 4.1: traversing the data to be classified, calculating the distance between the data and each clustering center according to the nearest matching principle, and dividing the data into the closest clustering centers according to the distance;
step 4.2: calculating the average value of each clustering center, taking the average value of the clustering centers as a new clustering center, judging whether a clustering function meets a convergence condition or the iteration times, taking the clustering result obtained in the step 4.1 as a sample characteristic if the clustering function meets the convergence condition or the iteration times, and repeating the step 4.1 if the clustering result does not meet the convergence condition or the iteration times;
step 4.3: and comparing Euclidean distances between the sample features to be classified and the sample features in the feature library to classify, and sorting the calculated Euclidean distances from small to large by calculating Euclidean distances between the feature components of the electrical appliance to be identified and the feature components in the feature library, wherein the feature to be identified is the nearest feature.
The collection device in step 1 is a non-invasive power load collection device of STM32, and the current characteristic is a discrete sequence.
The beneficial effects of the invention are as follows:
according to the non-invasive power load identification method, through collecting current signals or power signals of the electric appliance, the current characteristics of the electric appliance are decomposed and extracted by using an Empirical Mode Decomposition (EMD) method, the characteristics are optimized by using a k-means clustering method, and identification is performed by using a Euclidean distance nearest matching principle. The method solves the problem that in the prior art, when noise signals exist, the power load identification accuracy is not high. The problem of when the active reactive power of the electrical apparatus is close, discern the effect relatively poor is solved. The accuracy and the decomposition efficiency of the non-invasive power load decomposition are improved.
Drawings
FIG. 1 is a flow chart of a non-intrusive power load identification method of the present invention;
fig. 2 is a flow chart of decomposing the current characteristics of the electrical appliance using an empirical mode decomposition method.
Detailed Description
The invention will be described in detail below with reference to the drawings and the detailed description.
The invention discloses a non-invasive power load identification method, as shown in fig. 1, comprising the following steps:
step 1: collecting a current signal or a power meter signal of the electric appliance to obtain a current characteristic of the electric appliance, wherein the current characteristic is used as a sample characteristic in a characteristic library;
the collection device in the step 1 is a non-invasive power load collection device of STM32, and the current characteristic is a discrete sequence.
Step 2: as shown in fig. 2, decomposing the current characteristics of the electric appliance by using an empirical mode decomposition method to obtain sample characteristics to be classified;
step 2.1: determining the length L of the decomposed signal
The acquired signal defining a period is formed by at least 600 load data points, the length of the signal to be decomposed is at least the signal length of the period, the length L of the signal is selected, the current signal of the electric appliance with the length L is x (t), and in the embodiment, L=1000;
step 2.2: drawing an envelope curve of the signal according to the maximum value and the minimum value point of the signal, calculating the average value of the upper envelope curve and the lower envelope curve of the signal, and obtaining a signal h 1 (t): curve fitting is carried out on the signals x (t), and the upper envelope curves e of the signals are respectively fitted upp (t), lower envelope e low (t). Calculating average value of upper and lower envelope lines
Figure BDA0002433349440000051
And signal h 1 (t) is the signal to be decomposed x (t) minus m 1 And (t) the following: i.e. h 1 (t)＝x(t)-m 1 (t)；
Step 2.3: taking standard deviation of two successively screened decomposition signals as a judgment criterion of iteration stop, wherein the judgment criterion is defined as follows:
Figure BDA0002433349440000052
typically sd=0.2 to 0.3, h 1 (t) repeating step 2.2 as signal x (t) for k times until signal h is obtained 1k (t) stopping iteration if a criterion for stopping iteration is satisfied;
step 2.4: will signal h 1k (t) separating from the original signal: let c 1 (t)＝h 1k (t) defining a new variable c 1 (t) is the first basic feature component decomposed from the original signal, which contains the shortest periodic component in the original signal, and c 1 (t) separating from the original signal to obtain a signal r 1 (t)：r 1 (t)＝x(t)-c 1 (t); for the remaining signal component r 1 (t) performing EMD decomposition to obtain sample characteristics to be classified, namely:
Figure BDA0002433349440000053
r N (t) is the residual signal component obtained after N times of decomposition, the original current signal x (t) cannot be completely decomposed, r N The signal (t) may not be zero, and the number of feature values obtained may be used as a condition for terminating the decomposition. The invention obtains the first four decomposed signals c i (t) Current characteristics as an Electrical consumer, i.e. c 1 (t),c 2 (t),c 3 (t),c 4 (t) describes the current characteristics of the electrical consumer.
Step 3: optimizing the characteristics of the sample to be classified by adopting a k-means clustering method;
step 3.1: determining the number K of the categories of the clusters; k=2 n ；
Step 3.2: and arbitrarily selecting a group of load characteristics from each decomposed load sample as an initial center of the cluster.
Step 4: and identifying by using a Euclidean distance nearest matching principle.
Step 4.1: traversing the data to be classified, calculating the distance between the data and each clustering center according to the nearest matching principle, and dividing the data into the closest clustering centers according to the distance;
step 4.2: calculating the average value of each clustering center, taking the average value of the clustering centers as a new clustering center, judging whether a clustering function meets a convergence condition or the iteration times, if so, taking the clustering result obtained in the step 4.1 as a sample feature, otherwise, repeating the step 4.1, namely setting the initialized clustering center of the j-th sample feature as mu j Through the steps ofAfter step 4.1, classifying the sample feature set into the j-th class as Z j ，Z j ＝{z 1 ,z 2 ,z 3 ,...,z Y The number of the features is Y, z y For the feature signal after EMD decomposition, y= {1,2,3,..y }, the average value of the cluster centers is J j ：
Figure BDA0002433349440000061
Step 4.3: and comparing Euclidean distances between the sample features to be classified and the sample features in the feature library to classify, and sorting the calculated Euclidean distances from small to large by calculating Euclidean distances between the feature components of the electrical appliance to be identified and the feature components in the feature library, wherein the feature to be identified is the nearest feature.
The method adopts Euclidean distance between the sample characteristics to be classified and the sample characteristics in the characteristic library to classify, calculates Euclidean distance between the characteristic components of the electric appliance to be identified and the characteristic components in the characteristic library, and sorts the calculated Euclidean distance from small to large, and the characteristic to be identified is the nearest characteristic.
The data adopted in the embodiment is that the current data (or power data) of the electric appliance is collected by a non-invasive load collecting device. In the embodiment, four electric appliances, namely a computer, an induction cooker, a water boiling kettle and a blower, are used as target objects to verify the invention, wherein the current waveforms of the computer are complex, and the current waveforms of the induction cooker and the water boiling kettle are similar. Taking the length L=1000, namely taking the data length of 1000 sampling points as a study object, decomposing the signal through EMD, optimizing the characteristics by using k-means to establish a characteristic library, identifying by calculating Euclidean distance between the data to be identified and the data in the characteristic library, and finally identifying the data as shown in the table I.
Table one: recognition result
Figure BDA0002433349440000071
As can be seen from the identification results of the first and second tables, the method has a certain effect on the identification of the four electric appliances such as a computer, a kettle, an electromagnetic oven and a blower and the electric behavior combined with other electric appliances, and the identification accuracy is high. The method has universality for identifying other electric appliances. In general, the method effectively solves the problem in the prior art that the accuracy of identifying the power load is not high when noise signals and interference signals exist. The problem of when the active reactive power of the electrical apparatus is close, discern the effect relatively poor is solved. The accuracy and the decomposition efficiency of the non-invasive power load decomposition are improved.

Claims (5)

1. A method of non-intrusive power load identification, comprising the steps of:
step 1: collecting a current signal or a power meter signal of the electric appliance to obtain a current characteristic of the electric appliance, wherein the current characteristic is used as a sample characteristic in a characteristic library;
step 2: decomposing the current characteristics of the electric appliance by using an empirical mode decomposition method to obtain sample characteristics to be classified;
step 3: optimizing the characteristics of the sample to be classified by adopting a k-means clustering method;
step 4: and identifying by using a Euclidean distance nearest matching principle.
2. The non-invasive electrical load identification method according to claim 1, wherein the specific process of step 2 is:
step 2.1: determining the length L of the decomposed signal
Defining a period of acquisition signals by at least 600 load data points, wherein the length of a signal to be decomposed is at least the signal length of one period, the length L of the signal is selected, and the current signal of an electric appliance with the length L is x (t);
step 2.2: drawing an envelope curve of the signal according to the maximum value and the minimum value point of the signal, and obtaining the upper envelope curve and the lower envelope curve of the signalAverage value of (1) to obtain signal h 1 (t)；
Step 2.3: taking standard deviation of two successively screened decomposition signals as a judgment criterion of iteration stop, and taking h 1 (t) repeating step 2.2 as signal x (t) for k times until signal h is obtained 1k (t) stopping iteration if a criterion for stopping iteration is satisfied;
step 2.4: will signal h 1k (t) separating from the original signal, for the remaining signal component r 1 And (t) performing EMD decomposition to obtain the characteristics of the sample to be classified.
3. The non-invasive electrical load identification method according to claim 1, wherein the specific process of step 3 is:
step 3.1: determining the number K of the categories of the clusters; k=2 n ；
Step 3.2: and arbitrarily selecting a group of load characteristics from each decomposed load sample as an initial center of the cluster.
4. The non-invasive electrical load identification method according to claim 1, wherein the specific process of step 4 is:
step 4.1: traversing the data to be classified, calculating the distance between the data and each clustering center according to the nearest matching principle, and dividing the data into the closest clustering centers according to the distance;
step 4.2: calculating the average value of each clustering center, taking the average value of the clustering centers as a new clustering center, judging whether a clustering function meets a convergence condition or the iteration times, taking the clustering result obtained in the step 4.1 as a sample characteristic if the clustering function meets the convergence condition or the iteration times, and repeating the step 4.1 if the clustering result does not meet the convergence condition or the iteration times;
step 4.3: and comparing Euclidean distances between the sample features to be classified and the sample features in the feature library to classify, and sorting the calculated Euclidean distances from small to large by calculating Euclidean distances between the feature components of the electrical appliance to be identified and the feature components in the feature library, wherein the feature to be identified is the nearest feature.
5. A method of non-invasive electrical load identification as claimed in claim 1, wherein the collection device in step 1 is a STM32 non-invasive electrical load collection device, and the current signature is a discrete sequence.
CN202010243561.XA 2020-03-31 2020-03-31 Non-invasive power load identification method Active CN111327118B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010243561.XA CN111327118B (en) 2020-03-31 2020-03-31 Non-invasive power load identification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010243561.XA CN111327118B (en) 2020-03-31 2020-03-31 Non-invasive power load identification method

Publications (2)

Publication Number Publication Date
CN111327118A CN111327118A (en) 2020-06-23
CN111327118B true CN111327118B (en) 2023-07-07

Family

ID=71171770

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010243561.XA Active CN111327118B (en) 2020-03-31 2020-03-31 Non-invasive power load identification method

Country Status (1)

Country Link
CN (1) CN111327118B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113034305B (en) * 2021-02-10 2022-08-30 上海千居智科技有限公司 Non-invasive load monitoring event classification method and storage medium
CN113033688A (en) * 2021-04-06 2021-06-25 国网浙江省电力有限公司营销服务中心 Non-invasive load decomposition method
CN115051363B (en) * 2022-08-17 2023-01-13 广东电网有限责任公司佛山供电局 Distribution network area user change relation identification method and device and computer storage medium
CN117477771A (en) * 2023-10-30 2024-01-30 国网浙江省电力有限公司杭州市临平区供电公司 Non-invasive dormitory electricity safety monitoring method, system and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103839197A (en) * 2014-03-19 2014-06-04 国家电网公司 Method for judging abnormal electricity consumption behaviors of users based on EEMD method
WO2014165369A1 (en) * 2013-04-03 2014-10-09 The Gillette Company Electrochemical cell including an integrated circuit
CN109886498A (en) * 2019-03-01 2019-06-14 北京邮电大学 A kind of EMD-GRU Short-Term Load Forecasting Method based on feature selecting
CN110119545A (en) * 2019-04-24 2019-08-13 西安理工大学 A kind of non-intrusive electrical load recognition methods based on stack self-encoding encoder
CN110689190A (en) * 2019-09-24 2020-01-14 南方电网科学研究院有限责任公司 Power grid load prediction method and device and related equipment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014165369A1 (en) * 2013-04-03 2014-10-09 The Gillette Company Electrochemical cell including an integrated circuit
CN103839197A (en) * 2014-03-19 2014-06-04 国家电网公司 Method for judging abnormal electricity consumption behaviors of users based on EEMD method
CN109886498A (en) * 2019-03-01 2019-06-14 北京邮电大学 A kind of EMD-GRU Short-Term Load Forecasting Method based on feature selecting
CN110119545A (en) * 2019-04-24 2019-08-13 西安理工大学 A kind of non-intrusive electrical load recognition methods based on stack self-encoding encoder
CN110689190A (en) * 2019-09-24 2020-01-14 南方电网科学研究院有限责任公司 Power grid load prediction method and device and related equipment

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
A Hybrid FCW-EMD and KF-BA-SVM Based Model for Short-term Load Forecasting;Qingzhen Liu et al.;《CSEE JOURNAL OF POWER AND ENERGY SYSTEMS》;20180630;第4卷(第2期);第226-237页 *
一种基于双层聚类分析的负荷形态组合识别方法;王星华等;《电网技术》;20160531;第40卷(第5期);第1495-1501页 *
基于CEEMDAN和多尺度排列熵的球磨机负荷识别方法;胡显能等;《噪声与振动控制》;20180630;第38卷(第3期);第146-151页 *

Also Published As

Publication number Publication date
CN111327118A (en) 2020-06-23

Similar Documents

Publication Publication Date Title
CN111327118B (en) Non-invasive power load identification method
Lin et al. Development of an improved time–frequency analysis-based nonintrusive load monitor for load demand identification
CN109598451B (en) Non-invasive load identification method based on PCA (principal component analysis) and LSTM (least Square TM) neural network
CN103197218B (en) A kind of high-voltage cable insulation defect partial discharge electrification detection diagnostic method
CN107025365B (en) A kind of non-intruding load discrimination method for user side
CN111830347B (en) Two-stage non-invasive load monitoring method based on event
CN109858522A (en) A kind of management line loss abnormality recognition method based on data mining
CN111242391B (en) Machine learning model training method and system for power load identification
CN105823948A (en) Non-invasive resident load identification method
CN108197817A (en) A kind of method of the non-intrusion type load transient state monitoring based on big data
CN102566555B (en) Major appliance work state monitoring method based on pattern recognition
CN110119545B (en) Non-invasive power load identification method based on stack type self-encoder
CN113036759B (en) Fine granularity identification method and identification system for power consumer load
CN111553444A (en) Load identification method based on non-invasive load terminal data
CN111325485B (en) Light-weight gradient elevator power quality disturbance identification method considering internet-of-things bandwidth constraint
CN116629458B (en) Energy storage equipment operation optimization method and system based on data analysis
CN113887912B (en) Non-invasive load identification method for embedded equipment under deep learning
CN110084719A (en) A kind of distribution network load type device for identifying
CN103793756A (en) Transformer economic operation characteristic analyzing method
CN117559443A (en) Ordered power utilization control method for large industrial user cluster under peak load
CN112615428A (en) Line loss analysis and treatment system and method
CN111665388A (en) Non-invasive household air conditioner identification method
CN117543589A (en) Scheduling method of cascade hydropower safety centralized control system
CN112862163A (en) Short-term power load prediction system based on DTW algorithm
CN114239762A (en) Non-invasive load identification method and system based on structured load characteristic spectrum

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant