WO2022111741A1 - 一种云端协同的负荷辨识系统及方法 - Google Patents

一种云端协同的负荷辨识系统及方法 Download PDF

Info

Publication number
WO2022111741A1
WO2022111741A1 PCT/CN2022/075039 CN2022075039W WO2022111741A1 WO 2022111741 A1 WO2022111741 A1 WO 2022111741A1 CN 2022075039 W CN2022075039 W CN 2022075039W WO 2022111741 A1 WO2022111741 A1 WO 2022111741A1
Authority
WO
WIPO (PCT)
Prior art keywords
load
data
feature
characteristic
target
Prior art date
Application number
PCT/CN2022/075039
Other languages
English (en)
French (fr)
Inventor
刘兴奇
祝恩国
邹和平
林繁涛
雷民
徐英辉
陈昊
巫钟兴
张宇鹏
朱子旭
韩月
Original Assignee
中国电力科学研究院有限公司
国家电网有限公司
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 中国电力科学研究院有限公司, 国家电网有限公司 filed Critical 中国电力科学研究院有限公司
Priority to US17/759,769 priority Critical patent/US11841387B2/en
Publication of WO2022111741A1 publication Critical patent/WO2022111741A1/zh

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
    • G01R21/133—Arrangements for measuring electric power or power factor by using digital technique
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G01R19/25—Arrangements for measuring currents or voltages or for indicating presence or sign thereof using digital measurement techniques
    • G01R19/2513—Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/21—Design, administration or maintenance of databases
    • G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
    • G16Y10/00—Economic sectors
    • G16Y10/35—Utilities, e.g. electricity, gas or water
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Definitions

  • the present application relates to the technical field of load identification, and in particular, to a cloud-based collaborative load identification system and method.
  • the local load feature library cannot be updated in time after deployment, and can only identify parts. Electrical equipment, and when users access new equipment, they are often unable to identify correctly, which brings great limitations to the usage scenarios of the equipment. Therefore, it is necessary to consider adopting a method to solve the online identification problem of local load identification equipment.
  • the embodiments of the present application expect to provide a cloud-based collaborative load identification system and method.
  • an embodiment of the present application proposes a cloud-based collaborative load identification system, including:
  • the intelligent IoT electric energy meter module is used to identify the load characteristic data of the user's electrical equipment, and extract the characteristic quantity data from the load characteristic data with the preset load characteristic quantity extraction algorithm, and compare the extracted characteristic quantity data with the characteristic quantity data.
  • the first load characteristic database of the intelligent IoT electric energy meter module is matched, and the load characteristic data corresponding to the characteristic quantity data that has not been matched in the characteristic quantity data is determined as the target load characteristic data;
  • the master station load identification module is used to receive the target load characteristic data, perform data orientation processing on the target load characteristic data, obtain characteristic quantity data to be identified, and identify the characteristic quantity data from the to-be-identified characteristic quantity data.
  • Types of the target smart IoT power meter module and target load feature extraction algorithm in the case of determining that the target load feature extraction algorithm matches the target smart IoT power meter module, the matching information is stored and determined to be matched with the target load feature extraction algorithm.
  • the optimal matching strategy for matching the feature data to be identified according to the optimal matching strategy, identify the optimal matching solution corresponding to the feature data to be identified from the second load feature library, so as to complete the to-be-identified feature data.
  • the feature quantity data of , and the second load feature library are matched.
  • an embodiment of the present application also proposes a method for cloud-based collaborative load identification, including:
  • the intelligent IoT energy meter module identifies the load characteristic data of the user's electrical equipment, and extracts the characteristic quantity data from the load characteristic data with a preset load characteristic quantity extraction algorithm, and compares the extracted characteristic quantity data with the intelligent
  • the first load characteristic database of the IoT electric energy meter module performs matching, and determines the load characteristic data corresponding to the characteristic quantity data that has not been matched in the characteristic quantity data as the target load characteristic data;
  • the master station load identification module receives the target load characteristic data, performs data orientation processing on the target load characteristic data, obtains characteristic quantity data to be identified, and identifies the target intelligent object from the to-be-identified characteristic quantity data
  • the type of the connected energy meter module and the target load feature extraction algorithm when it is determined that the target load feature extraction algorithm matches the target smart IoT energy meter module, the matching information is stored and determined to be the same as the target load feature extraction algorithm. Identify the optimal matching strategy for feature data matching, and according to the optimal matching strategy, identify the optimal matching solution corresponding to the feature data to be identified from the second load feature library, so as to complete the feature to be identified. Matching of data to the second load signature library.
  • the intelligent IoT electric energy meter module is used to identify the load characteristic data of the user's electrical equipment, and extract the characteristic quantity data from the load characteristic data with the preset load characteristic quantity extraction algorithm, and compare the extracted characteristic quantity data with the characteristic quantity data.
  • the first load characteristic database of the intelligent IoT electric energy meter module is matched, and the load characteristic data corresponding to the characteristic quantity data that has not been matched in the characteristic quantity data is determined as the target load characteristic data;
  • the master station load identification module is used to receive the target load characteristic data, perform data orientation processing on the target load characteristic data, obtain characteristic quantity data to be identified, and identify the characteristic quantity data from the to-be-identified characteristic quantity data.
  • Types of the target smart IoT power meter module and target load feature extraction algorithm in the case of determining that the target load feature extraction algorithm matches the target smart IoT power meter module, the matching information is stored and determined to be matched with the target load feature extraction algorithm.
  • the optimal matching strategy for matching the feature data to be identified according to the optimal matching strategy, identify the optimal matching solution corresponding to the feature data to be identified from the second load feature library, so as to complete the to-be-identified feature data.
  • the feature quantity data of , and the second load feature library are matched.
  • the target load characteristic data is pre-/
  • the acquisition module is sent to the main station load identification module, and the target load characteristic data is identified through the main station load identification module.
  • the characteristic quantity data to be identified is obtained through the target load characteristic data, and then the corresponding characteristic quantity data to be identified is determined. the optimal matching solution. In this way, the function of online identification of unknown equipment by a load identification device such as a local intelligent IoT electric energy meter module can be realized, and the load identification capability of the equipment can be strengthened.
  • FIG. 1 is a schematic diagram of the composition and structure of a cloud-based collaborative load identification system according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of the composition and structure of another cloud-based collaborative load identification system provided by an embodiment of the present application;
  • FIG. 3 is a schematic diagram of the composition and structure of another cloud-based collaborative load identification system provided by an embodiment of the present application.
  • FIG. 4 is a schematic structural diagram of the composition of still another cloud-based collaborative load identification system according to an embodiment of the present application.
  • FIG. 5 is a schematic flowchart of the implementation of a method for cloud-based collaborative load identification provided by an embodiment of the present application
  • FIG. 6 is a schematic diagram of an implementation flowchart of a cloud-based collaborative load identification method provided by an embodiment of the present application.
  • the load identification of residential-side electrical equipment is an important direction of smart grid research.
  • the user's electrical equipment has the characteristics of a wide variety of equipment, large equipment scale, and large differences in the load characteristics of each equipment.
  • the mainstream electric energy meters installed on the user side only realize the measurement function of total electricity consumption, and do not realize the function of classifying electricity consumption according to the nature of electric equipment.
  • energy saving suggestions can be put forward for the power consumption period and power consumption composition of each electrical equipment, so that users can timely understand the energy consumption status of specific electrical equipment, so as to guide users to take energy saving measures independently.
  • the technology of resident user load identification has important research significance.
  • Load identification is generally divided into intrusive load identification and non-intrusive load identification technology.
  • Intrusive load identification technology requires separate measurement for each type of load or key electrical equipment of the user. This method requires the installation of a large number of hardware devices. Installation, maintenance and information processing consume a lot of costs, and the economic benefits are poor, which is an important reason for the difficulty of large-scale popularization.
  • the non-intrusive load identification method only needs to install the equipment with load identification at the user's power meter to collect power consumption data such as voltage and current, and perform load identification based on the local identification algorithm and load feature database, so as to analyze the household status of all loads within.
  • This method does not require hardware installation and maintenance, reduces costs, and is easier to implement.
  • the cloud-based collaborative load identification system 10 includes:
  • the intelligent IoT electric energy meter module 101 is used for identifying the load characteristic data of the user's electrical equipment, and extracting the characteristic quantity data for the load characteristic data with a preset load characteristic quantity extraction algorithm, and extracting the characteristic quantity data from the extracted load characteristic quantity data. Matching with the first load feature library of the smart IoT watt-hour meter module, and determining the load feature data corresponding to the unmatched feature data in the feature data as the target load feature data;
  • the master station load identification module 103 is configured to receive the target load characteristic data, perform data orientation processing on the target load characteristic data, obtain characteristic quantity data to be identified, and identify the characteristic quantity data to be identified from the characteristic quantity data to be identified. Determine the type of the target smart IoT power meter module and the target load feature extraction algorithm; in the case that the target load feature extraction algorithm is determined to match the target smart IoT power meter module, the matching information is stored and determined.
  • the optimal matching strategy matched with the feature data to be identified, according to the optimal matching strategy identify the optimal matching solution corresponding to the feature quantity data to be identified from the second load feature library, so as to complete the to-be-identified feature data. Matching of the identified feature quantity data with the second load feature library.
  • the smart IoT electric energy meter module 101 is a device with a load identification function installed at the user electric energy meter.
  • the master station load identification module 103 is a device or device for user load identification in the cloud.
  • the smart IoT power meter module 101 may include at least two non-invasive load identification modules, which have the function of identifying the load of the user's electrical equipment.
  • the first load characteristic library includes a personalized load characteristic library and a general load characteristic library; the general load characteristic library may include each non-intrusive load identification module in the smart IoT electric energy meter module 101 The sub-universal load feature library of the MCU; the personalized load feature library may include the sub-individualized load feature library of each non-intrusive load identification module in the smart IoT electric energy meter module 101; wherein, the general load feature library is used to record Feature data of at least two types of user electrical equipment; a personalized load feature library for recording feature data corresponding to the optimal matching solution.
  • the target load characteristic data may include unmatched data orientation processing on the target load characteristic data to obtain the characteristic quantity data to be identified, including at least the target energy connected energy meter corresponding to the matched characteristic quantity data.
  • Types of module and target load feature extraction algorithms are possible.
  • data orientation processing is performed on the target load characteristic data to obtain the characteristic quantity data to be identified, which may be the target load characteristic data to perform functions such as data cleaning and garbage screening, and the load characteristic quantity identification algorithm in the cloud can be used to identify the data. , and identify the feature data corresponding to the target load feature data as the feature data to be identified.
  • the target load characteristic data is pre-collected/collected by using the mining information.
  • the module is sent to the main station load identification module, and the target load characteristic data is identified through the main station load identification module.
  • the characteristic quantity data to be identified is obtained through the target load characteristic data, and then the corresponding characteristic quantity data to be identified is determined. optimal matching solution. In this way, the function of online identification of unknown equipment by a load identification device such as a local intelligent IoT electric energy meter module can be realized, and the load identification capability of the equipment can be strengthened.
  • the main station load identification module is also used to input the characteristic quantity data corresponding to the optimal matching solution after completing the matching of the characteristic quantity data to be identified and the second load characteristic library. the first load signature library.
  • the unknown equipment identified online can be synchronized to the load feature library in the equipment identified by the local user.
  • FIG. 2 is a schematic diagram of the composition and structure of another cloud-based collaborative load identification system provided by an embodiment of the present application.
  • the cloud-based collaborative load identification system 20 includes:
  • the intelligent IoT electric energy meter module 201 is used to identify the load characteristic data of the user's electrical equipment, and extract the characteristic quantity data for the load characteristic data with a preset load characteristic quantity extraction algorithm, and extract the characteristic quantity data for the extracted load characteristic quantity data. Matching with the first load feature library of the smart IoT watt-hour meter module, and determining the load feature data corresponding to the unmatched feature data in the feature data as the target load feature data;
  • a data import/export component 203 configured to receive the target load characteristic data
  • the module classification and verification component 204 is used to perform data orientation processing on the target load characteristic data, obtain characteristic quantity data to be identified, and identify the target intelligent IoT electric energy meter from the characteristic quantity data to be identified Module and target load feature extraction algorithm type;
  • the cloud identification component 205 is used to store the matching information and determine the matching information with the feature quantity data to be identified when it is determined that the target load feature extraction algorithm matches the target smart IoT energy meter module.
  • the optimal matching strategy according to the optimal matching strategy, identify the optimal matching solution corresponding to the feature data to be identified from the second load feature database, so as to complete the feature data to be identified Matching with the second load signature library.
  • the target load characteristic data is received through the data import/export component, and the target load characteristic data is subjected to data orientation processing through the module classification and verification component, so as to obtain the characteristic quantity data to be identified, and identify the target load characteristic data.
  • the target smart IoT power meter module and the target load feature extraction algorithm type, and finally through the cloud identification component in the case of determining that the target load feature extraction algorithm matches the target smart IoT power meter module, the matching Information is stored and the optimal matching strategy that matches the feature data to be identified is determined, and according to the optimal matching strategy, the corresponding feature data to be identified is identified from the second load feature database.
  • the optimal matching solution can be used to identify unknown equipment functions online through the load identification device in the cloud, which strengthens the load identification capability of the equipment.
  • FIG. 3 is a schematic diagram of the composition structure of another cloud-based collaborative load identification system provided by an embodiment of the present application.
  • the cloud-based collaborative load identification system 30 includes:
  • the first load feature library 301 is used to record the load feature data of at least two kinds of user electrical equipment and the feature data corresponding to the optimal matching solution;
  • the load processing unit 302 is used to collect and process the user-side power information data in real time, obtain the voltage and current change values of each user's electrical equipment, and determine the start-up time point and the transient stage, the stable state of the corresponding equipment according to the sampled voltage and current change values.
  • the time interval of the load in the state stage, and the load characteristic data of the user electrical equipment is identified according to the time interval;
  • the load identification unit 303 is configured to extract feature data in the load feature data based on the preset load feature extraction algorithm, match the extracted feature data with the first load feature library, and extract The load characteristic data corresponding to the unmatched characteristic quantity data determined in the characteristic quantity data is determined as the target load characteristic data;
  • the master station load identification module 305 is configured to receive the target load characteristic data, perform data orientation processing on the target load characteristic data, obtain characteristic quantity data to be identified, and identify the characteristic quantity data to be identified from the characteristic quantity data to be identified. Determine the type of the target smart IoT power meter module and the target load feature extraction algorithm; in the case that the target load feature extraction algorithm is determined to match the target smart IoT power meter module, the matching information is stored and determined.
  • the optimal matching strategy matched with the feature data to be identified, according to the optimal matching strategy identify the optimal matching solution corresponding to the feature quantity data to be identified from the second load feature library, so as to complete the to-be-identified feature data. Matching of the identified feature quantity data with the second load feature library.
  • the smart IoT electric energy meter module includes at least two non-intrusive load identification modules;
  • the personalized load feature library includes corresponding to each of the non-intrusive load identification modules at least two sub-personalized feature libraries;
  • the general load feature library includes at least two sub-universal feature libraries corresponding to each of the non-intrusive load identification modules;
  • Each of the sub-universal feature libraries is used to record the feature quantity data of at least one user electrical equipment under the corresponding non-intrusive load identification module;
  • Each of the sub-personalized feature libraries is used to selectively record feature quantity data corresponding to the optimal matching solution.
  • the load processing unit collects and processes the user-side power information data in real time, obtains the voltage and current change value of each user's electrical equipment, and determines the start-up time point and transient state of the corresponding equipment according to the sampled voltage and current change value.
  • the load time interval of the stage and the steady state stage the load characteristic data of the user electrical equipment is identified according to the time interval; the load characteristic quantity data of at least one user electrical equipment and the optimal matching solution are recorded through the first load characteristic database.
  • the corresponding feature data; finally, the feature data in the load feature data is extracted by the load identification unit, the extracted feature data is matched with the first load feature database, and the feature data determined in the feature data is matched.
  • the load characteristic data corresponding to the unmatched characteristic quantity data is determined as the target load characteristic data. In this way, each user can locally determine whether there is an unmatched or identified electrical device.
  • FIG. 4 is a schematic diagram of the composition structure of still another cloud-based collaborative load identification system provided by an embodiment of the present application.
  • the cloud-based collaborative load identification system includes: an intelligent IoT power meter module 40 , an intelligent IoT The electric energy meter 41, the use information pre/collection equipment 42 and the main station load identification module 43, wherein the intelligent IoT electric energy meter module 40 includes a non-intrusive load identification module 1 401, a non-intrusive load identification module 1 402 and non-intrusive load identification module n 403; the master station load identification module 43 includes a data import/export component 431, a module classification and verification component 432, a cloud identification component 433 and a cloud collaboration component 434;
  • the non-intrusive load identification module 1 401, the non-intrusive load identification module 1 402 and the non-intrusive load identification module n 403 have the same composition structure, here, only the non-intrusive load identification module 1 401 composition to describe;
  • the non-intrusive load identification module 1 401 includes a general load characteristic library 4011, a personalized load characteristic library 4012, a general load processing unit 4013 and a local load identification unit 4014;
  • the smart IoT power meter module 40 is connected to the smart IoT power meter 41 , and the smart IoT power meter module 40 obtains load characteristic data (voltage, voltage data) through the smart IoT power meter 41 , and obtains load characteristic data (voltage, voltage data) from the load Unrecognized (matched) load characteristic data is determined in the characteristic data, and the unrecognized load characteristic data is output to the smart IoT electricity meter 41;
  • the mining information pre-/collection device 42 calls the unidentified load characteristic data in the smart IoT electric energy meter 41, and transmits the unidentified load characteristic data to the main station load identification connected to the mining information pre-/collection device 42
  • the module 43 identifies the unrecognized load characteristic data through the main station load identification module 43 to obtain the optimal solution, and transmits the optimal solution in reverse to the user information pre-/collection device 42, /
  • the acquisition device 42 transmits it to the smart IoT power meter 41 ; the
  • the personalized load feature database 4012 is used to record the matched user's load feature data and device information
  • the general load processing unit 4013 is used to collect and process the power information data on the user side in real time, and according to the real-time sampled voltage and current change values of the equipment operation to determine the start time point of the equipment and the time interval of the load in the transient and steady state stages, and extract the general load characteristics data;
  • the local load identification unit 4013 has a load feature extraction algorithm, and extracts feature information according to the extracted local load identification unit, and matches the individual load signature database or the general load signature database.
  • the module classification and verification component 432 realizes the directional processing of the uploaded feature data, data cleaning, garbage screening and other functions, and identifies the local load identification algorithm corresponding to the current module of the smart IoT energy meter. Different algorithms have personalized data processing. Strategy.
  • the cloud identification component 433 performs polling processing for the algorithm set possessed by the master station. When a certain algorithm matches the identification device type with the load characteristic quantity, the current matching information is registered in the identification library. After completing the polling of the master station algorithm set, The cloud identification component identifies the optimal solution of the current identification library. The selection strategy of the optimal solution can be based on the algorithm priority in the algorithm library and the matching device type with the most solutions. The matched feature information is sent to the smart energy meter through the cloud collaboration component. , and update the energy meter personalized feature library.
  • Using the data import/export component 431 can realize the main station load identification module 42 to import and export feature quantity database data and update the load identification algorithm, so as to expand the load identification capability of the main station.
  • the acquisition information pre-/collection equipment 42 is a device and service that integrates communication network and data transmission. It can communicate with the acquisition information pre-service of the master station through networks such as Ethernet and 4G wireless network, and exchange load identification feature quantity data.
  • the smart IoT electric energy meter 41 is used to collect the voltage and current signals of the user's electrical equipment.
  • non-intrusive load identification module 1 401 due to the non-intrusive load identification module 1 401 or the non-intrusive load identification module 1 402, there are different non-intrusive load identification modules due to differences in load identification algorithms of different manufacturers. However, only one of the non-intrusive load identification modes is configured for each smart IoT energy meter.
  • the general load signature database 4011 is the solidified local signature database data, which records the characteristic quantity data of various equipments.
  • the general load signature database 4011 is generally fixed during the operation stage of the equipment. Due to the difference of the load identification algorithm among different equipment manufacturers, the general load signature is Libraries vary between devices from different manufacturers.
  • the personalized load feature library 4012 records the matched user's feature quantity and equipment information. After the device is installed on the user side, it will analyze the load characteristics of the user's electrical appliances and match the load identification feature quantity, and further match and identify the feature quantity and a specific local algorithm. Electrical equipment, current electrical equipment and feature quantities are stored in the personalized load feature library, so that after running for a period of time, the personalized load feature library 4012 can record the load identification feature library information of all devices of the home user.
  • the general load processing unit 4013 collects and processes the power information data on the user side in real time, and can judge the starting time point of the device and the time interval of the load in the transient and steady state according to the real-time sampling voltage and current change value of the device operation, and can extract general features amount to submit to the master station for collaborative identification.
  • the local load identification unit 4013 is a load feature extraction algorithm, and extracts the feature data according to the local load identification unit 4013, and matches the personalized load signature database 4012 or the general load signature database 4011, and can cooperate with the master station in the case of local matching failure. Process the feature quantity data of the matching equipment assisted by the master station, and enter the personalized load feature database after obtaining the relevant electrical equipment information of the feature quantity successfully matched by the master station.
  • the embodiment of the present application further provides a method for cloud-based collaborative load identification. As shown in FIG. 5 , the method includes:
  • Step S501 The intelligent IoT electric energy meter module identifies the load characteristic data of the user's electrical equipment, and extracts the load characteristic data with a preset load characteristic extraction algorithm, and compares the extracted characteristic data with the load characteristic data.
  • the first load characteristic database of the intelligent IoT electric energy meter module is matched, and the load characteristic data corresponding to the characteristic quantity data that has not been matched in the characteristic quantity data is determined as the target load characteristic data;
  • Step S502 call the target load characteristic data with the information pre-collection/collection module, and transmit the target load characteristic data to the master station load identification module;
  • Step S503 The master station load identification module receives the target load characteristic data, performs data orientation processing on the target load characteristic data, obtains characteristic quantity data to be identified, and identifies the characteristic quantity data from the to-be-identified characteristic quantity data.
  • Types of the target smart IoT power meter module and target load feature extraction algorithm in the case of determining that the target load feature extraction algorithm matches the target smart IoT power meter module, the matching information is stored and determined to be matched with the target load feature extraction algorithm.
  • the optimal matching strategy for matching the feature data to be identified according to the optimal matching strategy, identify the optimal matching solution corresponding to the feature data to be identified from the second load feature library, so as to complete the to-be-identified feature data.
  • the feature quantity data of , and the second load feature library are matched.
  • the embodiment of the present application further provides a cloud-based collaborative load identification method, the method comprising:
  • Step 60 The intelligent IoT electric energy meter module identifies the load characteristic data of the user's electrical equipment, and extracts the characteristic quantity data from the load characteristic data with a preset load characteristic quantity extraction algorithm, and compares the extracted characteristic quantity data with the load characteristic quantity data.
  • the first load characteristic database of the intelligent IoT electric energy meter module is matched, and the load characteristic data corresponding to the characteristic quantity data that has not been matched in the characteristic quantity data is determined as the target load characteristic data;
  • Step 61 call the target load characteristic data with the information pre-processing/collection module, and transmit the target load characteristic data to the master station load identification module;
  • Step 62 The master station load identification module receives the target load characteristic data, performs data orientation processing on the target load characteristic data, obtains characteristic quantity data to be identified, and identifies the characteristic quantity data from the to-be-identified characteristic quantity data.
  • Types of the target smart IoT power meter module and target load feature extraction algorithm in the case of determining that the target load feature extraction algorithm matches the target smart IoT power meter module, the matching information is stored and determined to be matched with the target load feature extraction algorithm.
  • the optimal matching strategy for matching the feature data to be identified according to the optimal matching strategy, identify the optimal matching solution corresponding to the feature data to be identified from the second load feature library, so as to complete the to-be-identified feature data.
  • the matching of the feature quantity data with the second load feature library
  • Step 64 Enter the feature quantity data corresponding to the optimal matching solution into the second load feature library.
  • the first load feature library includes a personalized load feature library and a general load feature library; the method further includes: the sub-personalized load feature library records the features corresponding to the optimal matching solution volume data.
  • FIG. 6 is a block diagram of the implementation of yet another method for cloud-based collaborative load identification provided by an embodiment of the present application, as shown in FIG. 6 , including:
  • Step S601 local load sampling
  • the local load sampling may be that the smart IoT energy meter module collects load characteristic data of local user electrical equipment.
  • Step S602 local identification algorithm processing
  • the local identification algorithm processing may be that the intelligent IoT electric energy meter module performs feature extraction on the load feature data with a preset load feature extraction algorithm to obtain feature quantity data.
  • Step S603 personalized feature library matching
  • the personalized load feature library matching may be to match the extracted feature quantity data with the personalized load feature library. If the matching is successful, go to step S615; if the matching fails, go to step S604;
  • Step S604 general load feature library matching
  • step S604 may be to match the extracted feature quantity data with the general load feature library. If the matching is successful, go to step S605; if the matching fails, go to step S606.
  • Step S605 import the personalized feature library
  • step S605 may be to import the extracted feature quantity data into the personalized feature library, and then enter step S615;
  • Step S606 processing by the general load processing unit
  • step S606 includes: the general-purpose load processing unit mainly analyzes and processes the data output by 601 according to the data requirements of the main station load identification module, and passes the identification (software identification, hardware identification) of the package module and the algorithm to the analyzed and processed data. to match the data requirements of S608 and S609.
  • Step S607 cloud collaboration component processing
  • the cloud coordination component processing may be that the cloud coordination component receives the load characteristic data, and performs directional processing on the load characteristic data to obtain the characteristic quantity data to be identified;
  • Step S608 module classification and verification processing
  • step S608 may be to identify the type of the target smart IoT power meter module and the target load feature extraction algorithm from the feature quantity data to be identified, and determine the target load feature extraction algorithm and the target smart IoT power Whether the table module matches, in the case of matching, go to step S609;
  • Step S609 algorithm library processing
  • step S609 may be to start processing the feature data to be identified by the algorithm library, that is, selecting one of the algorithms from the algorithm library to start processing the feature data.
  • Step S610 load feature library identification
  • step S610 may be to use the load feature library of the master station to identify the feature quantity data to be identified, and if the identification is successful, go to step S611;
  • Step S611 identification library information registration
  • step S611 includes registering the identification result identified by each algorithm in the algorithm library.
  • Step S612 the algorithm library polling is completed
  • step S612 includes judging that the polling of each algorithm in the algorithm library is completed, and entering step S613 when it is determined that each algorithm in the algorithm library is polled; In this case, the process proceeds to step S609.
  • Step S613 Identify the optimal solution selection of the library
  • step S613 may be to select the optimal solution from the identification results corresponding to each algorithm in the algorithm library.
  • Step S614 issue the module feature quantity
  • step S614 may be to deliver the selected optimal solution to the feature library of the module.
  • Step S615 End.
  • the embodiment of the present application realizes the function of online identification of unknown equipment by a load identification device such as a local intelligent IoT electric energy meter, and strengthens the load identification capability of the equipment;
  • the embodiment of the present application can reduce the performance and storage capacity requirements of the local terminal load identification device, effectively reduce the hardware cost, and bring considerable economic benefits when the number of devices is very large.
  • the master station in the embodiment of the present application collects a large amount of load data, which can more accurately carry out load identification learning and algorithm research and improvement, easily form a virtuous cycle, and enhance the load identification capability of the master station.
  • the embodiments of the present application may be provided as a method, a system, or a 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, etc.) having computer-usable program code embodied therein.
  • the solutions in the embodiments of the present application may be implemented in various computer languages, for example, the object-oriented programming language Java and the literal translation scripting language JavaScript, and the like.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions
  • the apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

Abstract

本申请实施例公开了一种云端协同的负荷辨识系统及方法，该系统，包括：智能物联电能表模组，用于对提取的特征量数据与智能物联电能表模组的第一负荷特征库进行匹配，将特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；用采信息前置/采集模组，用于调用目标负荷特征数据，并将目标负荷特征数据传输至主站负荷辨识模组；主站负荷辨识模组，用于接收目标负荷特征数据，对目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解。

Description

一种云端协同的负荷辨识系统及方法
相关申请的交叉引用
本申请基于申请号为202011395279.X、申请日为2020年11月30日、申请名称为“一种云端协同的负荷辨识系统及方法”的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及负荷辨识技术领域，尤其涉及一种云端协同的负荷辨识系统及方法。
背景技术
相关技术中，由于负荷识别的设备中本地负荷特征库能够识别的用户电器设备的种类是有限的，而实际中用户电气设备种类繁多，本地负荷特征库在部署后不能及时更新，只能识别部分电器设备，且当用户接入新设备后往往无法正确识别，这对设备的使用场景带来很大局限性，因此需要考虑采用一种方法来解决本地负荷识别设备的在线识别问题。
发明内容
本申请实施例期望提供一种云端协同的负荷辨识系统及方法。
第一方面，本申请实施例提出了一种云端协同的负荷辨识系统，包括：
智能物联电能表模组，用于识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
用采信息前置/采集模组，用于调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组；
所述主站负荷辨识模组，用于接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特 征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。第二方面，本申请实施例还提出了一种云端协同的负荷辨识的方法，包括：
智能物联电能表模组识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
用采信息前置/采集模组调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组；
所述主站负荷辨识模组接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。智能物联电能表模组，用于识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
用采信息前置/采集模组，用于调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组；
所述主站负荷辨识模组，用于接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征 量数据与所述第二负荷特征库的匹配。本申请实施例中，通过将智能物联电能表模组未完成匹配(无法识别)的特征量数据对应的负荷特征数据确定为目标负荷特征数据，将目标负荷特征数据通过用采信息前置/采集模组输至主站负荷辨识模组，通过主站负荷辨识模组对目标负荷特征数据进行识别，先通过目标负荷特征数据获取待识别的特征量数据，然后确定待识别的特征量数据对应的最优匹配解。如此，可以实现了本地智能物联电能表模组等负荷识别装置在线识别未知设备功能，强化了设备的负荷识别能力。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，这些附图示出了符合本申请的实施例，并与说明书一起用于说明本申请的技术方案。
图1为本申请实施例提供的一种云端协同的负荷辨识系统的组成结构示意图；
图2为本申请实施例提供的另一种云端协同的负荷辨识系统的组成结构示意图；
图3为本申请实施例提供的又一种云端协同的负荷辨识系统的组成结构示意图；
图4为本申请实施例提供的再一种云端协同的负荷辨识系统组成结构示意图；
图5为本申请实施例提供的一种云端协同的负荷辨识的方法的实现流程示意图；
图6为本申请实施例提供的一种云端协同的负荷辨识方法的实现流程示意图。
具体实施方式
现在参考附图介绍本发明的示例性实施方式，然而，本发明可以用许多不同的形式来实施，并且不局限于此处描述的实施例，提供这些实施例是为了详尽地且完全地公开本发明，并且向所属技术领域的技术人员充分传达本发明的范围。对于表示在附图中的示例性实施方式中的术语并不是对本发明的限定。在附图中，相同的单元/元件使用相同的附图标记。
除非另有说明，此处使用的术语(包括科技术语)对所属技术领域的技术人员具有通常的理解含义。另外，可以理解的是，以通常使用的词典限定的术语，应当被理解为与其相关领域的语境具有一致的含义，而不应该被理解为理想化的或过于正式的意义。
居民侧用电设备负荷识别是智能电网研究的一个重要方向，用户的用电设备存在种类繁多、设备规模庞大、各设备负荷特征差异大等特点。目前在用户侧安装的主流电能表只实现了总用电量的计量功能，并未实现按用电设备性质进行分类电量计量功能。当 能够分析出家庭详细的能效构成信息，为各个用电设备的用电时段及用电量构成提出节能建议，使用户及时了解具体用电设备的用能状况，从而引导用户自主采取节能措施，为用户降低电费支出，进而有效降低能源的消耗和不合理浪费，实现节能降耗效果，因此居民用户负荷识别技术具有重要的研究意义。
负荷识别一般分为侵入式负荷识别和非侵入式负荷识别技术，侵入式负荷识别技术，需要针对用户的每类负荷或者重点用电设备进行单独计量，这种方式需要安装大量的硬件设备，在安装、维护及信息处理等方面需消耗大量的成本，经济效益差，这是难以大规模推广普及的重要原因。
非侵入式负荷识别方法，只需在用户的电力总表处安装带负荷识别的设备进行用电数据如电压、电流的采集，基于本地的辨识算法与负荷特征库进行负荷辨识，从而分析出家庭内所有负荷的状态。此方式无需硬件的安装及维护、降低了成本、更易于实现。
目前随着居民用户的负荷识别试点及推广，已经有不少本地侧负荷识别的设备在居民用户家庭部署，在实际使用过程中发现负荷识别设备一般要事先经过样本数据训练或学习过程；部分算法效率低，负荷辨识的实时性难以保证；由于电气设备种类繁多，工况复杂，比较难找到一种算法精确识别各个电器等情况。由于设备自身性能有限，本地负荷特征库在部署后不能及时更新，只能识别部分电器设备，且当用户接入新设备后往往无法正确识别，这对设备的使用场景带来很大局限性，因此需要考虑采用一种方法来解决本地负荷识别设备的在线识别及同步问题。
基于上述技术问题，本申请实施例提出了一种云端协同的负荷辨识系统，如图1所示，该云端协同的负荷辨识系统10包括：
智能物联电能表模组101，用于识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
用采信息前置/采集模组102，用于调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组103；
所述主站负荷辨识模组103，用于接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与 所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。
可以理解的是，智能物联电能表模组101是安装在用户电能表处的带负荷识别功能的设备。主站负荷辨识模组103是处于云端的用户负荷识别的设备或装置。
在一些可能的实施方式中，智能物联电能表模组101可以包括至少两个非侵入式负荷辨识模组，具备识别出用户电器设备负荷功能。
在一种实施方式中，第一负荷特征库，包括个性化负荷特征库和通用负荷特征库；通用负荷特征库可以包括智能物联电能表模组101中每一非侵入式负荷辨识模组中的子通用负荷特征库；个性化负荷特征库可以包括智能物联电能表模组101中每一非侵入式负荷辨识模组的子个性化负荷特征库；其中，通用负荷特征库，用于记录至少两种用户电器设备的特征量数据；个性化负荷特征库，用于记录所述最优匹配解对应的特征量数据。
可以理解的是，目标负荷特征数据可以包含未匹配的对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，至少包括完成匹配的特征量数据对应的目标能物联电能表模组和目标负荷特征提取算法的类型。
在一些实施方式中，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，可以是目标负荷特征数据进行数据清洗、垃圾筛选等功能，并通过云端的负荷特征量识别算法，识别出目标负荷特征数据对应的特征量数据作为待识别的特征量数据。
申请实施例中，通过将智能物联电能表模组未完成匹配(无法识别)的特征量数据对应的负荷特征数据确定为目标负荷特征数据，将目标负荷特征数据通过用采信息前置/采集模组输至主站负荷辨识模组，通过主站负荷辨识模组对目标负荷特征数据进行识别，先通过目标负荷特征数据获取待识别的特征量数据，然后确定待识别的特征量数据对应的最优匹配解。如此，可以实现了本地智能物联电能表模组等负荷识别装置在线识别未知设备功能，强化了设备的负荷识别能力。
可以理解的是，主站负荷辨识模组，还用于在完成所述待识别的特征量数据与所述第二负荷特征库的匹配之后，将所述最优匹配解对应的特征量数据录入所述第一负荷特征库。
可以看出，将最优匹配解对应的特征量数据录入所述第一负荷特征库，可以将在线 识别的未知设备同步到本地用户识别的设备中的负荷特征库。
图2为本申请实施例提供的另一种云端协同的负荷辨识系统的组成结构示意图，如图2所示，该云端协同的负荷辨识系统20包括：
智能物联电能表模组201，用于识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
用采信息前置/采集模组202，用于调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至数据导入/导出组件203；
数据导入/导出组件203，用于接收所述目标负荷特征数据；
模组分类与验证组件204，用于对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别的特征量数据中识别出所述目标智能物联电能表模组和目标负荷特征提取算法类型；
云端辨识组件205，用于在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别的特征量数据匹配的所述最优匹配策略，根据所述最优匹配策略，从所述第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。
本申请实施例中，通过数据导入/导出组件接收所述目标负荷特征数据，通过模组分类与验证组件对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，并识别出所述目标智能物联电能表模组和目标负荷特征提取算法类型，最后通过云端辨识组件在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别的特征量数据匹配的所述最优匹配策略，根据所述最优匹配策略，从所述第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，从而可以通过云端的负荷识别装置在线识别未知设备功能，强化了设备的负荷识别能力。
图3为本申请实施例提供的又一种云端协同的负荷辨识系统的组成结构示意图，如图3所示，该云端协同的负荷辨识系统30包括：
第一负荷特征库301，用于记录至少两种用户电器设备的负荷特征量数据和所述最优匹配解对应的特征量数据；
负荷处理单元302，用于实时采集处理用户侧电力信息数据，得到每一用户电器设备的电压电流变化值，根据所述采样电压电流变化值来判断对应设备的开启时间点及暂态阶段、稳态阶段的负荷的时间区间，根据所述时间区间识别用户电器设备的负荷特征数据；
负荷辨识单元303，用于基于预设的所述负荷特征量提取算法提取所述负荷特征数据中的特征量数据，对提取的所述特征量数据与所述第一负荷特征库进行匹配，将所述特征量数据中确定未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
用采信息前置/采集模组304，用于调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组305；
所述主站负荷辨识模组305，用于接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。
在一些可能的实施方式中，所述智能物联电能表模组包括至少两个非介入式负荷辨识模组；所述个性化负荷特征库包括与每一所述非介入式负荷辨识模组对应的至少两个子个性化特征库；所述通用负荷特征库包括与每一所述非介入式负荷辨识模组对应的至少两个子通用特征库；
每一所述子通用特征库，用于记录对应的所述非介入式负荷辨识模组下的至少一种用户电器设备的特征量数据；
每一所述子个性化特征库，用于选择性记录所述最优匹配解对应的特征量数据。
本申请实施例中，通过负荷处理单元实时采集处理用户侧电力信息数据，得到每一用户电器设备的电压电流变化值，根据所述采样电压电流变化值来判断对应设备的开启时间点及暂态阶段、稳态阶段的负荷的时间区间，根据所述时间区间识别用户电器设备的负荷特征数据；通过第一负荷特征库记录至少一种用户电器设备的负荷特征 量数据和所述最优匹配解对应的特征量数据；最后通过负荷辨识单元提取所述负荷特征数据中的特征量数据，对提取的所述特征量数据与所述第一负荷特征库进行匹配，将所述特征量数据中确定未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据。如此每一用户可以在本地确定出自身是否存在未匹配或识别的电器设备。
图4为本申请实施例提供的再一种云端协同的负荷辨识系统的组成结构示意图，如图4所示，该云端协同的负荷辨识系统包括：智能物联电能表模组40、智能物联电能表41、用采信息前置/采集设备42和主站负荷辨识模组43，其中，智能物联电能表模组40包括非介入式负荷辨识模组1 401、非介入式负荷辨识模组1 402和非介入式负荷辨识模组n 403；主站负荷辨识模组43包括数据导入/导出组件431、模组分类与验证组件432、云端辨识组件433和云端协同组件434；
非介入式负荷辨识模组1 401、非介入式负荷辨识模组1 402和非介入式负荷辨识模组n 403的组成结构是相同的，这里，仅对非介入式负荷辨识模组1 401的组成进行描述；
非介入式负荷辨识模组1 401包括通用负荷特征库4011、个性化负荷特征库4012、通用负荷处理单元4013和本地负荷辨识单元4014；
图4中，智能物联电能表模组40与智能物联电能表41相连，智能物联电能表模组40通过智能物联电能表41获取负荷特征数据(电压、电压数据)，并从负荷特征数据中确定出未识别(匹配)的负荷特征数据，并将未识别的负荷特征数据输出给智能物联电表41；智能物联电能表41与用采信息前置/采集设备42连接，用采信息前置/采集设备42调用智能物联电能表41中的未识别的负荷特征数据，并将未识别的负荷特征数据传输至与用采信息前置/采集设备42连接的主站负荷辨识模组43，通过主站负荷辨识模组43识别未识别的负荷特征数据，得到最优解，并将最优解反向传递给用采信息前置/采集设备42，通过用采信息前置/采集设备42传递给智能物联电能表41；智能物联电能表模组40获取智能物联电能表41中的最优解，将最优解对应的特征量存储到智能物联电能表41中对应的特征库中。本申请实施例中，通用负荷特征库4011是固化的本地特征库信息，用来记录多种用户电器设备的负荷特征数据；
个性化负荷特征库4012用于记录匹配的用户的负荷特征数据及设备信息；
通用负荷处理单元4013用于实时采集处理用户侧电力信息数据，根据设备运行的实时采样电压电流变化值来判断设备的开启时间点及暂态及稳态阶段的负荷的时间区 间，提取通用负荷特征数据；
本地负荷辨识单元4013具有负荷特征量提取算法，并依据提本地负荷辨识单元取特征量信息，并匹配个性化负荷特征库或通用负荷特征库。模组分类与验证组件432实现主站对上传特征量数据定向处理，数据清洗、垃圾筛选等功能，识别智能物联电能表当前模组对应的本地负荷识别算法，不同算法具备个性化的数据处理策略。
云端辨识组件433针对主站具备的算法集进行轮询处理，当某种算法具备与负荷特征量匹配识别设备类型情况下，当前匹配信息登记于识别库中，完成轮询主站算法集后，云端辨识组件识别出目前识别库的最优解，最优解的选择策略可基于算法库中算法优先级以及匹配设备类型最多解等方法，匹配的特征量信息通过云端协同组件下发智能电能表，并更新电能表个性化特征库。
采用数据导入/导出组件431可实现主站负荷辨识模组42导入及导出特征量库数据及更新负荷识别算法，扩展主站的负荷识别能力。
用采信息前置/采集设备42是集成通讯网络及数据传输的设备及服务，可通过以太网、4G无线网等网络与主站的采集信息前置服务通讯，交换负荷识别特征量数据。
智能物联电能表41用于采集用户电器设备的电压电流信号。
其中，由于非侵入式负荷辨识模组1 401或非介入式负荷辨识模组1 402因为不同厂商的负荷识别算法的差异存在不同的非侵入式负荷辨识模组。但对于每一智能物联电能表只配置其中一种非侵入式负荷辨识模。
通用负荷特征库4011是固化的本地特征库数据，记录多种设备的特征量数据，在设备运行阶段通用负荷特征库4011一般固定不变，不同设备厂商间因负荷识别算法的差异因此通用负荷特征库在不同厂商设备间存在差异性。
个性化负荷特征库4012是记录匹配的用户的特征量及设备信息，在设备安装在用户侧后会分析用户电器的负荷特征并匹配负荷识别特征量，进一步通过特征量及特定的本地算法匹配识别用电设备，当前的用电设备及特征量存储于个性化负荷特征库中，这样当运行一段时间后个性化负荷特征库4012可记录家庭用户所有设备的负荷识别特征库信息。
通用负荷处理单元4013是实时采集处理用户侧电力信息数据，可根据设备运行的实时采样电压电流变化值来判断设备的开启时间点及暂态及稳态阶段的负荷的时间区间，可提取通用特征量以提交主站协同识别。
本地负荷辨识单元4013是负荷特征量提取算法，并依据本地负荷辨识单元4013提 取特征量数据，并匹配个性化负荷特征库4012或通用负荷特征库4011，在本地匹配失败情况下可与主站协同处理由主站协助匹配设备的特征量数据，在获取主站成功匹配的特征量的相关用电设备信息后录入个性化负荷特征库。
在上述实施例的基础上，本申请实施例还提供了一种云端协同的负荷辨识的方法，如图5所示，该方法包括：
步骤S501：智能物联电能表模组识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
步骤S502：用采信息前置/采集模组调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组；
步骤S503：所述主站负荷辨识模组接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。
本申请实施例再提供了一种云端协同的负荷辨识的方法，该方法包括：
步骤60：智能物联电能表模组识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
步骤61：用采信息前置/采集模组调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组；
步骤62：所述主站负荷辨识模组接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所 述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配；
步骤64：将所述最优匹配解对应的特征量数据录入所述第二负荷特征库。
在一些实施方式中，所述第一负荷特征库，包括个性化负荷特征库和通用负荷特征库；所述方法还包括：所述子个性化负荷特征库记录所述最优匹配解对应的特征量数据。
图6为本申请实施例提供的再一种云端协同的负荷辨识的方法的实现流程框图，如图6所示，包括：
步骤S601：本地负荷采样；
这里，本地负荷采样可以是智能物联电能表模组采集本地用户电器设备的负荷特征数据。
步骤S602：本地辨识算法处理；
这里，本地辨识算法处理，可以是智能物联电能表模组对负荷特征数据以预设的负荷特征提取算法进行特征提取，得到特征量数据。
步骤S603：个性化特征库匹配；
这里，个性化负荷特征库匹配可以是对提取的特征量数据与个性化负荷特征库进行匹配，在匹配成功的情况下，进入步骤S615；在匹配失败的情况下，进入步骤S604；
步骤S604：通用负荷特征库匹配；
这里，步骤S604可以是对提取的特征量数据与通用负荷特征库进行匹配，在匹配成功的情况下，进入步骤S605；在匹配失败的情况下，进入步骤S606。
步骤S605：导入个性化特征库；
这里，步骤S605可以是将提取的特征量数据导入个性化特征库，然后进入步骤S615；
步骤S606：通用负荷处理单元处理；
这里，步骤S606包括：通用负荷处理单元主要依据主站负荷辨识模块的数据需求，分析处理601输出的数据，并对分析处理后的数据通过封装模组和算法的标识(软件标识、硬件标识)以匹配S608和S609的数据需求。
步骤S607：云端协同组件处理；
这里，云端协同组件处理可以是云端协同组件接收负荷特征数据，并对负荷特征数据进行定向处理，得到待识别的特征量数据；
步骤S608：模组分类与验证处理；
这里，步骤S608可以是从待识别的特征量数据中识别处于目标智能物联电能表模组和目标负荷特征提取算法的类型，并确定所述目标负荷特征提取算法与所述目标智能物联电能表模组是否匹配，在匹配的情况下，进入步骤S609；
步骤S609：算法库处理；
这里，步骤S609可以是对待识别的特征量数据开始进行算法库处理，即，从算法库中选择其中一个算法对特征量数据开始处理。
步骤S610：负荷特征库识别；
这里，步骤S610可以是使用主站的负荷特征库对待识别的特征量数据进行识别，在识别成功的情况下，进入步骤S611；
步骤S611：识别库信息登记；
这里，步骤S611包括登记算法库中每一算法识别的识别结果。
步骤S612：算法库轮询完毕；
这里，步骤S612包括判断算法库中的每一算法轮询完毕，在确定算法库中的每一算法轮询完毕的情况下进入步骤S613；在确定算法库中的每一算法轮询未完成的情况下进入步骤S609。
步骤S613：识别库最优解选择；
这里，步骤S613可以是从算法库中每一算法对应的识别结果中选择最优解。
步骤S614：下发模组特征量；
这里，步骤S614可以是将选择的最优解下发给模组的特征库。
步骤S615：结束。
本申请实施例实现了本地智能物联电能表等负荷识别装置在线识别未知设备功能，强化了设备的负荷识别能力；
本申请实施例基于云端强大的计算能力，能降低等本地端负荷识别设备的性能和存储容量要求，有效降低硬件成本，当设备量非常大的情况下将带来可观的经济效益。
本申请实施例主站集中了海量的负荷数据，能够更加精准的开展负荷识别学习和算法的研究及改进，易形成良性循环，增强主站负荷识别能力。
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的 计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。本申请实施例中的方案可以采用各种计算机语言实现，例如，面向对象的程序设计语言Java和直译式脚本语言JavaScript等。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本申请的优选实施例，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例作出另外的变更和修改。所以，所附权利要求意欲解释为包括优选实施例以及落入本申请范围的所有变更和修改。
显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。

Claims (9)

  1. 一种云端协同的负荷辨识系统，所述系统包括：
    智能物联电能表模组，用于识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
    用采信息前置/采集模组，用于调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组；
    所述主站负荷辨识模组，用于接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别的特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。
  2. 根据权利要求1所述的系统，所述主站负荷辨识模组，还用于在所述完成所述待识别的特征量数据与所述第二负荷特征库的匹配之后，将所述最优匹配解对应的特征量数据录入所述第二负荷特征库。
  3. 根据权利要求1或2所述的系统，其中，所述第一负荷特征库，包括个性化负荷特征库和通用负荷特征库；
    所述通用负荷特征库，用于记录至少两种用户电器设备的特征量数据；
    所述个性化负荷特征库，用于记录所述最优匹配解对应的特征量数据。
  4. 据权利要求1或2所述的系统，其中，所述主站负荷辨识模组，包括：
    数据导入/导出组件，用于接收所述目标负荷特征数据；
    模组分类与验证组件，用于对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别的特征量数据中识别出所述目标智能物联电能表模组和目标负荷特征提取算法类型；
    云端辨识组件，用于在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别的特征量数据匹配的 所述最优匹配策略，根据所述最优匹配策略，从所述第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配；
    对应地，所述用采信息前置/采集模组，用于调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至所述数据导入/导出组件。
  5. 根据权利要求3所述的系统，其中，所述智能物联电能表模组，包括：
    所述第一负荷特征库，用于记录至少两种用户电器设备的负荷特征量数据和所述最优匹配解对应的特征量数据；
    负荷处理单元，用于实时采集处理用户侧电力信息数据，得到每一用户电器设备的电压电流变化值，根据所述采样电压电流变化值来判断对应设备的开启时间点及暂态阶段、稳态阶段的负荷的时间区间，根据所述时间区间识别用户电器设备的负荷特征数据；
    负荷辨识单元，用于基于预设的所述负荷特征量提取算法提取所述负荷特征数据中的特征量数据，对提取的所述特征量数据与所述第一负荷特征库进行匹配，将所述特征量数据中确定未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据。
  6. 据权利要求5所述的系统，其中，所述智能物联电能表模组，包括至少两个非介入式负荷辨识模组；所述个性化负荷特征库包括与每一所述非介入式负荷辨识模组对应的至少两个子个性化特征库；所述通用负荷特征库包括与每一所述非介入式负荷辨识模组对应的至少两个子通用特征库；
    每一所述子通用特征库，用于记录对应的所述非介入式负荷辨识模组下的至少两种用户电器设备的特征量数据；
    每一所述子个性化特征库，用于选择性记录所述最优匹配解对应的特征量数据。。
  7. 一种云端协同的负荷辨识方法，所述方法包括：
    智能物联电能表模组识别用户电器设备的负荷特征数据，并对所述负荷特征数据以预设的负荷特征量提取算法进行特征量数据提取，对提取的所述特征量数据与所述智能物联电能表模组的第一负荷特征库进行匹配，将所述特征量数据中未完成匹配的特征量数据对应的负荷特征数据确定为目标负荷特征数据；
    用采信息前置/采集模组调用所述目标负荷特征数据，并将所述目标负荷特征数据传输至主站负荷辨识模组；
    所述主站负荷辨识模组接收所述目标负荷特征数据，对所述目标负荷特征数据进行数据定向处理，得到待识别的特征量数据，从所述待识别特征量数据中识别出目标智能物联电能表模组和目标负荷特征提取算法的类型；在确定所述目标负荷特征提取算法与所述目标智能物联电能表模组匹配的情况下，对匹配信息进行存储并确定与所述待识别特征数据匹配的最优匹配策略，根据所述最优匹配策略，从第二负荷特征库中识别所述待识别的特征量数据对应的最优匹配解，以完成所述待识别的特征量数据与所述第二负荷特征库的匹配。
  8. 据权利要求7所述的方法，在所述主站负荷辨识模组完成所述待识别的特征量数据与所述第二负荷特征库的匹配之后，将所述最优匹配解对应的特征量数据录入所述第二负荷特征库。
  9. 据权利要求7或8所述的方法，所述第一负荷特征库，包括个性化负荷特征库和通用负荷特征库；所述方法还包括：
    所述子个性化负荷特征库记录所述最优匹配解对应的特征量数据。
PCT/CN2022/075039 2020-11-30 2022-01-29 一种云端协同的负荷辨识系统及方法 WO2022111741A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/759,769 US11841387B2 (en) 2020-11-30 2022-01-29 Cloud-end collaborative system and method for load identification

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011395279.X 2020-11-30
CN202011395279.XA CN112560908A (zh) 2020-11-30 2020-11-30 一种云端协同的负荷辨识系统及方法

Publications (1)

Publication Number Publication Date
WO2022111741A1 true WO2022111741A1 (zh) 2022-06-02

Family

ID=75047404

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/075039 WO2022111741A1 (zh) 2020-11-30 2022-01-29 一种云端协同的负荷辨识系统及方法

Country Status (3)

Country Link
US (1) US11841387B2 (zh)
CN (1) CN112560908A (zh)
WO (1) WO2022111741A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112560908A (zh) * 2020-11-30 2021-03-26 中国电力科学研究院有限公司 一种云端协同的负荷辨识系统及方法

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105186693A (zh) * 2015-09-28 2015-12-23 南方电网科学研究院有限责任公司 一种非侵入式用电负荷辨识系统及方法
CN106124852A (zh) * 2016-06-13 2016-11-16 国网天津宝坻供电有限公司 非侵入式负荷监测系统
CN109001528A (zh) * 2018-06-14 2018-12-14 国网江苏省电力有限公司电力科学研究院 一种非侵入式分项计量采集系统
CN109870600A (zh) * 2019-04-04 2019-06-11 江苏林洋能源股份有限公司 一种具有自学习能力的非侵入式负荷辨识方法以及智能电表
CN110672934A (zh) * 2019-08-23 2020-01-10 北京中电飞华通信股份有限公司 一种非侵入式负荷辨识方法、终端及系统
US20200311598A1 (en) * 2019-03-28 2020-10-01 International Business Machines Corporation Exposing payload data from non-integrated machine learning systems
CN111830345A (zh) * 2020-07-11 2020-10-27 中国电力科学研究院有限公司 一种非介入式负荷辨识测试方法及系统
CN112560908A (zh) * 2020-11-30 2021-03-26 中国电力科学研究院有限公司 一种云端协同的负荷辨识系统及方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112051462A (zh) * 2020-07-24 2020-12-08 南京航灵信息科技有限公司 基于大数据的负荷分解实现方法及系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105186693A (zh) * 2015-09-28 2015-12-23 南方电网科学研究院有限责任公司 一种非侵入式用电负荷辨识系统及方法
CN106124852A (zh) * 2016-06-13 2016-11-16 国网天津宝坻供电有限公司 非侵入式负荷监测系统
CN109001528A (zh) * 2018-06-14 2018-12-14 国网江苏省电力有限公司电力科学研究院 一种非侵入式分项计量采集系统
US20200311598A1 (en) * 2019-03-28 2020-10-01 International Business Machines Corporation Exposing payload data from non-integrated machine learning systems
CN109870600A (zh) * 2019-04-04 2019-06-11 江苏林洋能源股份有限公司 一种具有自学习能力的非侵入式负荷辨识方法以及智能电表
CN110672934A (zh) * 2019-08-23 2020-01-10 北京中电飞华通信股份有限公司 一种非侵入式负荷辨识方法、终端及系统
CN111830345A (zh) * 2020-07-11 2020-10-27 中国电力科学研究院有限公司 一种非介入式负荷辨识测试方法及系统
CN112560908A (zh) * 2020-11-30 2021-03-26 中国电力科学研究院有限公司 一种云端协同的负荷辨识系统及方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
TIAN, XIN: "Development and Application of New Intelligent Ammeter based on Multi-layer Cooperative Load Identification Technology", POWER SYSTEMS AND BIG DATA, vol. 23, no. 8, 1 August 2020 (2020-08-01), pages 18 - 25, XP055936405, DOI: 10.19317/j.cnki.1008-083x.2020.08.003 *
ZEIFMAN, MICHAEL ET AL.: "Nonintrusive Appliance Load Monitoring: Review and Outlook", "IEEE TRANSACTIONS ON CONSUMER ELECTRONICS", vol. 57, no. 1, 21 February 2011 (2011-02-21), XP011478006, DOI: 10.1109/TCE.2011.5735484 *

Also Published As

Publication number Publication date
US20230070173A1 (en) 2023-03-09
CN112560908A (zh) 2021-03-26
US11841387B2 (en) 2023-12-12

Similar Documents

Publication Publication Date Title
CN110350528B (zh) 一种低压台区拓扑自动辨识方法
CN112510841B (zh) 配电自动化的拓扑识别方法及装置
CN106443244B (zh) 用电设备类型辨识方法和系统
CN111242391B (zh) 用于电力负荷识别的机器学习模型训练方法及系统
CN107180314B (zh) 一种基于一二次系统关联关系的运维业务模型建模方法
CN103105556A (zh) 基于稳态与暂态特征联合匹配的智能电网负载检测与识别方法
CN104635114A (zh) 一种电能质量扰动源定位系统及定位方法
WO2022111741A1 (zh) 一种云端协同的负荷辨识系统及方法
CN106093565A (zh) 一种基于稳态特征波形匹配的电能分项计量方法及装置
CN112804111B (zh) 一种具有台区拓扑识别功能的4g通信单元及识别方法
CN114977176B (zh) 电力负荷分解方法、装置、设备及存储介质
CN110968703B (zh) 基于lstm端到端抽取算法的异常计量点知识库构建方法及系统
CN113887912A (zh) 一种深度学习下面向嵌入式设备的非侵入式负荷识别方法
CN102323975B (zh) 基于iec61850模型文件的报文正确性判别方法
CN112615428A (zh) 一种线损分析治理系统及方法
CN112182499B (zh) 一种基于时序电量数据的低压配电网拓扑结构辨识方法
CN110719445A (zh) 一种基于图像识别的远程抄表系统及方法
CN111178679A (zh) 一种基于聚类算法和网络搜索的相位识别方法
CN115186935A (zh) 一种机电设备非线性故障预测方法及系统
Yu et al. Practical method for data-driven user phase identification in low-voltage distribution networks
CN112051462A (zh) 基于大数据的负荷分解实现方法及系统
CN113985266B (zh) 多智能电表集中化负荷识别方法
CN114200203B (zh) 一种变电站计量装置故障监控方法和系统
CN113723844B (zh) 一种基于集成学习的低压台区理论线损计算方法
CN109002683B (zh) 社会网络发现算法验证系统及方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22725971

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 19/09/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 22725971

Country of ref document: EP

Kind code of ref document: A1