WO2020215636A1 - 非侵入终端辨识能力测试案例库构建方法及模拟检测平台 - Google Patents

非侵入终端辨识能力测试案例库构建方法及模拟检测平台 Download PDF

Info

Publication number
WO2020215636A1
WO2020215636A1 PCT/CN2019/113106 CN2019113106W WO2020215636A1 WO 2020215636 A1 WO2020215636 A1 WO 2020215636A1 CN 2019113106 W CN2019113106 W CN 2019113106W WO 2020215636 A1 WO2020215636 A1 WO 2020215636A1
Authority
WO
WIPO (PCT)
Prior art keywords
case
intrusive
electrical
current
voltage
Prior art date
Application number
PCT/CN2019/113106
Other languages
English (en)
French (fr)
Inventor
邓士伟
傅萌
苗青
黄莉
丁皓
戴聪
刘海杰
Original Assignee
江苏智臻能源科技有限公司
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 江苏智臻能源科技有限公司 filed Critical 江苏智臻能源科技有限公司
Publication of WO2020215636A1 publication Critical patent/WO2020215636A1/zh

Links

Classifications

    • 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
    • G06F16/22—Indexing; Data structures therefor; Storage structures
    • G06F16/2282—Tablespace storage structures; Management thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • 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
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • the invention relates to a method for constructing a test case library of non-invasive terminal identification capabilities and a simulation detection platform thereof, and belongs to the technical field of intelligent electricity use.
  • Non-intrusive load identification technology refers to a technology that analyzes the detailed usage of household appliances in the user through the user port voltage and current data, so as to provide basic data for two-way interaction of power users such as demand response, smart power consumption, and integrated energy services. Compared with the intrusive sub-measurement method, the investment is small and practical, and the application prospect is broad.
  • Non-intrusive terminal functions include data collection, load identification and data upload.
  • the testing content covers type test testing, general function testing and core identification function testing.
  • Existing detection methods have reference significance for type test detection and general function detection of non-invasive terminals, but they cannot be applied to the test of the core function of non-invasive terminals-load identification ability.
  • Aiming at the non-intrusive terminal identification ability test one of the core functions of the simulation test system is to simulate and provide real residential electricity usage scenarios.
  • the patent of the present invention proposes a residential electricity usage scenario construction plan, which represents the electrical operation scene as a single case. Build a case library.
  • the purpose of the present invention is to provide a method for constructing a non-invasive terminal identification capability test case library and a simulation detection platform to solve the above-mentioned problems in the background art.
  • a method for constructing a test case library for non-intrusive terminal identification capabilities includes the following steps:
  • Step 1 Basic research: select an area for establishing a database, collect data in this area, collect basic information of household appliances and the operation of each appliance, and make records;
  • Step 2 Case design: build a case model based on the basic information of household appliances collected in step one; build a case library framework based on the operating conditions of household appliances collected in step one;
  • Step 3 Case library construction: use case recording method and case storage method to complete the construction of the case library.
  • the basic information includes the category of electrical appliances, the name of the electrical appliance, the brand of the electrical appliance, and the operating principle of the electrical appliance.
  • the operating principle of the electrical appliance includes the type of electrical appliance, the power model of the electrical appliance, and the functional mode of the electrical appliance.
  • the overall framework of the case library includes four groups of level1, level2, level3, and level4.
  • the level1 group includes several cases, and each case runs in a single electrical appliance in a single mode, superimposing background electrical appliances;
  • the level2 group It includes several cases, each of which is two electrical appliances in two modes, staggered start-up time, superimposed operation, and superimposed background appliances;
  • the level3 includes several cases, each case has three electrical appliances in three modes, staggered start-up time , Superimpose operation, superimpose background appliances;
  • the level4 group includes several cases, each case has four modes of four appliances, staggered start-up time, superimposed operation, superimposed background appliances; the superimposition methods are in line with the collected appliances happening.
  • the case model of level1 includes 15 numerical attribute variables; among the 15 attribute variables, L1-L2 are the level attributes and total running time of the case, and L3-L5 are the electrical appliances category and name And electrical appliance brands, the three-dimensional information is uniformly numbered with all electrical appliances as a major category, and the value is the corresponding number value; L6 is the principle type of electrical appliance, the number is the combination of the appliance number and the type number, and the value is the parallel combination of two numbers; L7 is Function operation mode, the number is the combination of the appliance number and the function number, and the value is the parallel combination of two numbers; L 8 is the operating power value in the case scenario; L9-L12 are the operating time, operating calendar, season and weather temperature, corresponding to each Record time, date, season number and temperature; L13-L14 are load current and load voltage, which are represented by current time series value and voltage time series value respectively; L15 is identification result series value, used for comparison and analysis of non-intrusive terminal identification results ;
  • the Level2-level4 case model contains the same 15 numerical attribute variables as Leve1.
  • the attributes L3-L8 are multi-dimensional feature vectors, and the dimension is the number of superimposed electrical operating conditions.
  • the level2 case is 2-dimensional.
  • the case of level3 is 3-dimensional; other attribute information is consistent with the corresponding attribute information of level1.
  • the case recording method includes the following steps
  • level1 For the level1 case, set the operating scene according to its L 1 - L 8 , start the corresponding electrical appliance operation, record the voltage and current at the gate as the L 13 - L 14 sequence value, and divide the current and voltage of the electrical appliance at the same time Monitor and count the electrical name, power, and breaking time, and the result that needs to be identified is used as the L 15 sequence value;
  • L 3 - L 8 For level2-level4 cases, set the operating scenarios according to L 3 - L 8. According to the staggered start time in the collected electrical operating conditions, turn on and off the corresponding electrical appliances in turn, and record the voltage and current at the gate as L 13 - L 14 sequence value, while monitoring the current and voltage of the sub-item electrical appliances, and calculate the name, power, and breaking time of each sub-item electrical appliance. The identification result is used as the L 15 sequence value.
  • the design of the case storage method includes the following steps
  • the non-intrusive terminal simulation detection platform includes a test host, a serial port server, a non-intrusive identification terminal, and a high-current waveform playback device.
  • the test host is connected through the serial server and the non-intrusive identification terminal.
  • the high-current waveform playback device is equipped with Ethernet Interface, voltage output port, and current output port.
  • the Ethernet interface realizes signal interconnection with the test host, the voltage output port is connected to a non-intrusive identification terminal through a voltage transformer, and the current output port is connected to a non-intrusive identification terminal through a current transformer.
  • the test host receives the case library, the test host receives the case library information, and the result of identifying the non-intrusive identification terminal, and generates a report; the test host outputs the voltage and current waveforms of the test case to the large current waveform playback instrument, and the waveform playback The instrument reproduces the port voltage and current of the real user.
  • the non-intrusive identification terminal is provided with a 485 serial port, which is connected to a serial server; the working voltage of the voltage transformer is 220V, and the working current of the current transformer is 100A.
  • the test host outputs the voltage and current waveforms ( L 13 - L 14 ) of the test case to the large current waveform playback instrument, the waveform playback instrument reproduces the port voltage and current that simulates the real user, and the non-intrusive terminal passes through the current transformer and the voltage interface Collect voltage and current, start the identification module for load identification, and send the identification result to the test host through the serial server, and the test host compares the identification result with the real result ( L 15 ) for evaluation.
  • the present invention simulates one of the core functions of the test system for non-intrusive terminal identification ability testing, and simulates the provision of real residential electricity use scenarios, namely, proposes a construction plan for residential electricity use scenarios, and expresses the electrical operation scenarios as a single case.
  • Case Library The patented test case library of the present invention is applied to a non-intrusive terminal simulation detection platform, and is a maximum simulation of the complex and changeable operation scenes of residential household appliances.
  • Figure 1 is a schematic flow diagram of the present invention
  • Figure 2 is a schematic diagram of the structure of the case library framework
  • Figure 3 is a schematic diagram of the structure of a non-intrusive terminal simulation detection platform.
  • FIG. 1 is a schematic diagram of the structure of the present invention. It can be seen from the accompanying drawings that this method for constructing a test case library for non-invasive terminal identification capabilities is characterized by including the following steps:
  • Step 1 Basic research: select an area for establishing a database, collect data in this area, collect basic information of household appliances and the operation of each appliance, and make records;
  • Step 2 Case library design: build a case model based on the basic information of household appliances collected in step one; build a case library framework based on the operating conditions of household appliances collected in step one;
  • Step 3 Case library design: use case recording method and case storage method to complete the construction of the case library.
  • the non-intrusive terminal is installed at the main opening of the building meter box to collect the three-phase voltage at the branch point and the user current at each user entry switch; identify the user's electrical and non-electrical characteristics according to the user port voltage and current characteristics, and then Identify the type of electrical load within the user and realize the refined perception of the load within the user. Identification capability is the key core function of non-intrusive terminals, which characterizes the accuracy of the identification of electrical appliances and power consumption by non-intrusive terminals.
  • the basic information of electrical appliances includes the survey of the composition of household appliances of residential users, including the categories of appliances, the name of the appliance, and the brand of the appliance; the survey of the operating principle of electrical appliances, that is, the investigation of the difference in external operating characteristics caused by the principle of electrical appliances, such as the type of appliance and the power of the appliance. Model, electrical function mode, etc.
  • Research methods for investigating basic information include household surveys, online e-commerce sales data surveys, and offline physical store sales data surveys, and use the data of existing non-intrusive measurement model equations to assist statistics, such as the Jiangsu 863 Suzhou pilot project.
  • the electrical appliance types are based on the principle of covering most electrical appliance types, and the survey information is statistically analyzed.
  • the electrical appliances are mainly medium and high power appliances.
  • the household appliances are classified and sorted, and the appliances are classified into air-conditioning, electric heating, electric heating, kitchen, other and background categories, and the residential users under the general categories are sorted
  • the names, models, function modes and brands of common medium and high-power electrical appliances are selected.
  • the brands selected top3 brands, and the market coverage of top3 brands has been investigated.
  • the results of electrical appliances sorting are shown in Table 3.
  • each type of electrical appliance has a power model distinction.
  • the power model is selected according to the power range of the same product. Table 3 is the result of the investigation.
  • the operation scene investigation of household appliances is the investigation of usage habits, including the regularity of general electrical appliances superimposed use, the time regularity of electrical appliance use, the regularity of opening time interval of electrical superimposed operation, and the air-conditioning usage law, which are indirect manifestations of user behavior.
  • the survey methods of household appliances operation scenes include: household survey methods are used to conduct user behavior and habits surveys, and data from existing non-intrusive measurement model equations are used to assist statistics. Based on the principle of covering the vast majority of electrical appliance operating scenarios, the survey information is statistically analyzed.
  • the single device scene in Table 3 is superimposed on background appliances.
  • the background appliances are always on, refrigerators and lights, and TV sets and laptops are randomly added.
  • the running time of the single-device scenario case refers to 1/3 of the normal running time of the appliance but covers all modes of the working process. For appliances that are used for a long time such as air conditioners, choose 20 minutes.
  • the electric appliance superimposed operation scenarios in Table 3, and the electric appliance superimposing method is determined through research, select high power consumption (over 1000kWh in winter or summer 4 months) and normal power consumption (winter or winter). In summer 4 months, the electricity consumption is 200kWh-1000kWh) each 50 households are investigated and statistically analyzed.
  • the main difference between the high power consumption in winter and summer is the amount and time of air conditioners.
  • the superimposition model mainly focuses on multiple air conditioners, which are superimposed on air conditioners and other appliances.
  • the superposition ratio of air conditioners is greater than 50%. It is superimposed with other appliances by about 35%, and other appliances by themselves by about 15%; the superposition model in spring and autumn will be simpler, mainly the superposition of electric heating equipment and kitchen appliances.
  • the case library framework includes four groups of level1, level2, level3, and level4.
  • the level1 group includes several cases, each of which runs in a single appliance and single mode, superimposed on background appliances;
  • the level2 group includes several cases, each of which is Two electrical appliances in two modes, staggered start-up time, superimposed operation, superimposed background appliances;
  • level3 includes several cases, each of which is three electrical appliances in three modes, staggered start-up time, superimposed operation, superimposed background appliances;
  • level4 includes There are several cases, each of which is four electrical appliances in four modes, staggered start time, superimposed operation, superimposed background appliances; superimposition methods are in line with the collected electrical operation conditions.
  • a case is a single operating scenario, also called a test case.
  • the case library is a collection of cases and a collection of electrical operating scenarios; the design principle of the case library is to follow objectivity, minimize a single case, and make the entire case library comprehensive; four Each group divides the operating scenarios from simple to complex, representing four operating scenarios with different levels of complexity, and simulates complex real power scenarios through the combination of cases.
  • the case model of level1 is shown in Table 1.
  • the 15 attribute variables are all numerical variables, among which L1- L2 are the level attributes and total running time of the case, and the basic attribute information of L3- L5 electrical appliances. This three-dimensional information is based on all electrical appliances.
  • the class is uniformly numbered, and the value is the corresponding number value;
  • the L6 appliance type is the external attribute feature reflecting the working principle of the appliance, and belongs to the appliance name, so the number is the combination of the appliance number and the type number, and the value is the parallel combination of two numbers; similarly L7 functional mode has similar attributes;
  • L 8 is the operating power value in the case scenario;
  • L9-L12 External meteorological factors, calendar factors, time factors, and seasonal factors are supplementary information for the case to aid identification.
  • L13- L14 are the voltage and current time series values of the electrical appliance operation, the main part of the case, the main basis for non-intrusive terminal identification, and L15 is the identification result sequence value, which is used for comparison and analysis of the results of non-intrusive terminal identification.
  • Level2-level4 is a scene mode in which two or more electrical appliances are superimposed and operated.
  • the characteristic attributes of multiple appliances involved in the case are numerical vectors composed of multiple elements.
  • the level2-level4 case model contains the same 15 numerical attribute variables as Leve1, among which the attributes L3-L8 are multi-dimensional feature vectors, and the dimension is the number of superimposed electrical operating conditions.
  • the level2 case is 2-dimensional, and the level3 case The case is three-dimensional; other attribute information is consistent with the corresponding attribute information of level 1.
  • the case recording method includes the following steps
  • level1 For the level1 case, set the operating scene according to its L 1 - L 8 , start the corresponding electrical appliance operation, record the voltage and current at the gate as the L 13 - L 14 sequence value, and divide the current and voltage of the electrical appliance at the same time Monitor and count the electrical name, power, and breaking time, and the result that needs to be identified is used as the L 15 sequence value;
  • L 3 - L 8 For level2-level4 cases, set the operating scenarios according to L 3 - L 8. According to the staggered start time in the collected electrical operating conditions, turn on and off the corresponding electrical appliances in turn, and record the voltage and current at the gate as L 13 - L 14 sequence value, while monitoring the current and voltage of the sub-item electrical appliances, and calculate the name, power, and breaking time of each sub-item electrical appliance. The identification result is used as the L 15 sequence value.
  • the case storage method design includes the following steps
  • the non-intrusive terminal simulation detection platform is characterized by including a test host, a serial server, a non-intrusive identification terminal, and a high-current waveform playback instrument.
  • the test host is connected through the serial server and the non-intrusive identification terminal.
  • the high-current waveform playback instrument is equipped with an Ethernet interface , Voltage output port and current output port, Ethernet interface and test host realize signal interconnection, voltage output port is connected to non-intrusive identification terminal through voltage transformer, and current output port is connected to non-intrusive identification terminal through current transformer.
  • the test host receives the case library, and the test host receives the case library information, as well as the result of identifying the non-intrusive identification terminal, and generates a report.
  • the test host outputs the voltage and current waveforms (L13-L14) of the test case to the large current waveform playback instrument.
  • the waveform playback instrument reproduces the port voltage and current that simulates the real user.
  • the non-intrusive terminal collects the voltage and current through the current transformer and the voltage interface. Start the identification module for load identification, and send the identification result to the test host through the serial server. The test host compares the identification result with the real result (L15) for evaluation.
  • the non-intrusive identification terminal is equipped with a 485 serial port, which is connected to a serial server; the working voltage of the voltage transformer is 220V, and the working current of the current transformer is 100A. There are several non-intrusive identification terminals.

Abstract

本发明公开了一种非侵入终端辨识能力的测试案例库构建方法，属于智能用电技术领域。其包括以下操作步骤：步骤1：基础调研：选定建立数据库的区域，在该区域内进行数据采集，采集家用电器的基本信息以及各电器的运行情况，并做记录；步骤2：案例设计：根据步骤一采集的家用电器基本信息建立案例的模型；根据步骤一采集的家电的运行情况建立案例库框架；步骤3：案例库设计：使用案例录波方法和案例入库方法完成案例库的构建。本发明提出一种居民用电场景的建设方案，将电器运行场景以单个案例表示，建设案例库。本发明专利测试案例库应用于非侵入终端模拟检测平台，是对复杂多变的居民用户家用电器运行场景的最大化模拟。

Description

非侵入终端辨识能力测试案例库构建方法及模拟检测平台 技术领域
本发明涉及一种非侵入终端辨识能力的测试案例库构建方法及其模拟检测平台，属于智能用电技术领域。
 背景技术
非侵入负荷辨识技术是指通过用户端口电压、电流数据分析用户内家用电器详细使用情况的技术，从而为需求响应、智能用电、能源综合服务等电力用户双向互动提供基础数据。对比于侵入式的分项计量方式，投入小实用性强，应用前景广阔。
面向非侵入终端的大规模工程应用需求，非侵入终端的检测标准规范和辨识功能的检测能力建设却稍显滞后。当前中国电科院、中国电机工程学会等相关机构正组织制定相关标准。非侵入终端的检测能力亟需建设，江苏电科院、江苏智臻能源科技有限公司等建设的实证环境，主要服务于非侵入式负荷辨识技术的研究试验，无法满足非侵入终端的批量化检测需求。
非侵入终端功能包括数据采集、负荷辨识和数据上传，检测内容涵盖型式试验检测、一般功能检测和核心辨识功能检测。已有检测方法对于非侵入终端的型式试验检测、一般功能检测具有参考意义，但无法应用于非侵入终端的核心功能-负荷辨识能力的测试。针对非侵入终端辨识能力测试，模拟测试系统的核心功能之一，模拟提供真实的居民用电场景，本发明专利即提出一种居民用电场景的建设方案，将电器运行场景以单个案例表示，建设案例库。
 技术问题
本发明的目的在于提供非侵入终端辨识能力测试案例库构建方法及模拟检测平台，以解决上述背景技术中提出的问题。
 技术解决方案
一种非侵入终端辨识能力的测试案例库构建方法，包括以下操作步骤：
步骤1：基础调研：选定建立数据库的区域，在该区域内进行数据采集，采集家用电器的基本信息以及各电器的运行情况，并做记录；
步骤2：案例设计：根据步骤一采集的家用电器基本信息建立案例的模型；根据步骤一采集的家电的运行情况建立案例库框架；
步骤3：案例库建设：使用案例录波方法和案例入库方法完成案例库的构建。
更进一步的，所述基本信息包括电器大类、电器名称、电器品牌和电器运行原理，电器运行原理包括电器类型、电器功率型号和电器功能模式，基本信息调研手段选用入户调研、线上电商销售数据调研、线下实体店销售数据调研和利用已有的非侵入量测示范方程的数据；所述各电器的运行情况包括电器叠加使用规律、电器使用的时间规律和电器叠加运行的开启时间间隔规律，电器运行情况的调研手段选用入户调研和利用已有的非侵入量测示范方程的数据。
更进一步的，所述案例库整体框架包括level1、level2、level3和level4四组，所述level1组中包括若干个案例，每个案例为单台电器单模式运行，叠加背景电器；所述level2组中包括若干个案例，每个案例为两台电器两种模式，错开启动时间，叠加运行，叠加背景电器；所述level3包括若干个案例，每个案例为三台电器三种模式，错开启动时间，叠加运行，叠加背景电器；所述level4组中包括若干个案例，每个案例为四台电器四种模式，错开启动时间，叠加运行，叠加背景电器；所述叠加方式均符合采集的电器运行情况。
更进一步的，所述level1的案例模型包括15个数值型的属性变量；所述15个属性变量中，L1- L2为案例的级别属性和总运行时长，L3- L5为电器大类、电器名称和电器品牌，所述三维信息以所有电器为大类进行统一编号，数值为对应编号值；L6为电器原理类型，编号为电器编号与类型编号的组合，数值为两个编号并列组合；L7为功能运行模式，编号为电器编号与功能编号的组合，数值为两个编号并列组合；L 8为案例场景中的运行功率值；L9- L12为运行时刻、运行日历、季节和天气温度，分别对应记录时间、日期、季节编号和温度；L13- L14为负荷电流和负荷电压，分别用电流时间序列值和电压时间序列值表示；L15为辨识结果序列值，用于非侵入终端辨识的结果对比分析；
Level2- level4的案例模型包含与Leve1相同的15个数值型的属性变量，其中属性L3- L8为多维数特征向量，维数即为叠加的电器工况数，所述level2的案例是2维，level3的案例是3维；其他属性信息则与level1的相应的属性信息表述一致。
更进一步的，所述案例录波方法包括如下步骤
1）选取录制地点：选取装有侵入和非侵入量测系统的试点示范家庭、对比验证平台为场所进行录制；
2）对于level1案例，按照其 L 1- L 8设定运行场景，启动对应的电器运行，录制关口处的电压及电流，作为 L 13 - L 14 序列值，同时进行分项电器的电流、电压监测，统计出电器名称、电量、和开断时间，需要辨识的结果，作为 L 15 序列值；
3）对于level2- level4案例，按照 L 3 - L 8 设定运行场景，根据采集的电器的运行情况中的错开的启动时间，依次启动及关闭对应的电器，录制关口处的电压及电流，作为 L 13 - L 14 序列值，同时进行分项电器的电流、电压监测，统计出各分项电器名称、电量、开断时间等，需要辨识的结果，作为 L 15 序列值。
更进一步的，所述案例入库方法设计，包括如下步骤
1）提炼案例模型的属性信息构建案例模型表结构；
2）根据案例模型表结构构建对应的EXCEL文档结构；
3）将步骤5得到的案例波形及属性信息进行EXCEL中录入，批量化导入系统数据库；
4）根据案例模型表结构开发案例互动化前台界面，实现案例的增、删、改、查。
非侵入终端模拟检测平台，包括测试主机、串口服务器、非侵入辨识终端和大电流波形回放仪，所述测试主机通过串口服务器和非侵入辨识终端连接，所述大电流波形回放仪设置有以太网接口、电压输出端口和电流输出端口，所述以太网接口与测试主机实现信号互联，所述电压输出端口通过电压互感器与非侵入辨识终端连接，所述电流输出端口通过电流互感器与非侵入辨识终端连接；所述测试主机接收案例库，测试主机接收案例库信息，以及辨识非侵入辨识终端的结果，并生成报告；测试主机输出测试案例的电压电流波形至大电流波形回放仪，波形回放仪复现模拟真实用户的端口电压电流。
更进一步的，所述非侵入辨识终端设置有485串口，所述485串口与串口服务器连接；所述电压互感器的工作电压为220V，电流互感器的工作电流为100A。
更进一步的，所述非侵入辨识终端有若干个。
更进一步的，测试主机输出测试案例的电压电流波形（ L 13 - L 14 ）至大电流波形回放仪，波形回放仪复现模拟真实用户的端口电压电流，非侵入终端通过电流互感器及电压接口采集电压、电流，并启动辨识模块进行负荷辨识，将辨识结果通过串口服务器上送至测试主机，测试主机将辨识结果与真实结果（ L 15 ）进行比对评价。
有益效果
本发明针对非侵入终端辨识能力测试，模拟测试系统的核心功能之一，模拟提供真实的居民用电场景，即提出一种居民用电场景的建设方案，将电器运行场景以单个案例表示，建设案例库。本发明专利测试案例库应用于非侵入终端模拟检测平台，是对复杂多变的居民用户家用电器运行场景的最大化模拟。
 附图说明
图1是本发明的流程示意图；
图2是案例库框架的结构示意图，
图3是非侵入终端模拟检测平台的结构示意图。
 本发明的实施方式
下面结合附图和具体实施方式，进一步阐明本发明。应理解下述具体实施方式仅用于说明本发明而不用于限制本发明的范围。
图1是本发明的结构示意图，结合附图可见，本种非侵入终端辨识能力的测试案例库构建方法，其特征在于包括以下操作步骤：
步骤1：基础调研：选定建立数据库的区域，在该区域内进行数据采集，采集家用电器的基本信息以及各电器的运行情况，并做记录；
步骤2：案例库设计：根据步骤一采集的家用电器基本信息建立案例的模型；根据步骤一采集的家电的运行情况建立案例库框架；
步骤3：案例库设计：使用案例录波方法和案例入库方法完成案例库的构建。
非侵入终端安装于楼宇表箱总开处，采集分支点处三相电压及各用户入户开关处采集用户电流；根据用户端口电压、电流特性辨识用户的电气特征量和非电气特征量，进而辨识用户内电器负荷的类型，实现用户内负荷的精细化感知。辨识能力是非侵入终端的关键核心功能，表征非侵入终端对电器类型和消耗电量辨识的准确性。
电器的基本信息包括即是调研居民用户家用电器的构成，包含电器大类，电器名称，电器品牌；调研电器运行原理，即电器原理构成导致的外在运行特性差异调研，如电器类型、电器功率型号、电器功能模式等。调研基本信息的调研手段包括入户调研、线上电商销售数据调研和线下实体店销售数据调研，并且利用已有的非侵入量测示范方程的数据辅助统计，如江苏863苏州试点工程。考虑到实际建设的可行性，电器类型以涵盖绝大多数电器类型为原则，对调研信息进行统计分析，另外考虑到负荷辨识技术现状及电网供需互动需求，电器类型主要是中大功率电器。
按照覆盖目前70%以上家庭用电场景为原则，进行了家用电器归类整理，将电器归为空调类、电热类、电暖类、厨房类、其他类及背景类，整理大类下居民用户常见中大功率电器名称、型号、功能模式及品牌，品牌挑选top3的品牌，并调研了top3品牌的市场覆盖率，电器整理结果如表3所示。除了表中的参数，此外每类电器都有功率型号区分，功率型号按同产品功率区间选取小中大三档。表3是调研梳理结果。
家用电器运行场景调研，即是使用习惯调研，包括一般电器叠加使用规律、电器使用的时间规律、电器叠加运行的开启时间间隔规律和空调使用规律，是用户行为的间接体现。
家用电器运行场景调研手段包括：采用入户调研方式进行用户行为习惯调研，并且利用已有的非侵入量测示范方程的数据辅助统计。以涵盖绝大多数电器运行场景为原则，将调研信息进行统计分析。
结合用户电器使用行为习惯分析，进行案例场景的分析设计：
Level1案例，表3中单设备场景，同时叠加背景类电器，背景电器电冰箱和电灯常开、电视机和笔记本随机加入。单设备场景案例运行时长参照该电器正常运行时长的1/3但涵盖工作过程所有模式，空调等长时间使用的电器则选择20分钟。
Level2-Level4案例，表3中电器叠加运行场景，电器叠加方式通过调研确定，从苏州试点用户中挑选高用电量（冬或夏季4个月用电超过1000kWh）及正常用电量（冬或夏季4个月用电在200kWh-1000kWh）用户各50户进行调研及统计分析。冬夏两季用电量高主要区别在于空调使用的数量与时间上，在这两季中叠加模型主要集中在多台空调，空调与其他电器的叠加，其中空调自身叠加的比例大于50%，空调与其他电器叠加约35%，其他电器自身叠加约15%；春秋两季叠加模型会更简单，主要就是电热类设备与厨房电器类的叠加。
以家用电器运行场景的调研结果为依据，进行案例库框架设计。案例库整体框架包括level1、level2、level3和level4四组，level1组中包括若干个案例，每个案例为单台电器单模式运行，叠加背景电器；level2组中包括若干个案例，每个案例为两台电器两种模式，错开启动时间，叠加运行，叠加背景电器；level3包括若干个案例，每个案例为三台电器三种模式，错开启动时间，叠加运行，叠加背景电器；level4组中包括若干个案例，每个案例为四台电器四种模式，错开启动时间，叠加运行，叠加背景电器；叠加方式均符合采集的电器运行情况。
其中，案例即是某单个运行场景，也叫测试案例，案例库是案例的集合，是电器运行场景的汇总；案例库设计原则为遵循客观性，单个案例最小化，整个案例库全面化；四个组将运行场景由简到繁划分，分别代表四个不同等级复杂程度的运行场景，通过案例的组合模拟复杂的真实用电场景。
level1的案例模型如表1所示，15个属性变量均为数值型变量，其中L1- L2为案例的级别属性和总运行时长，L3- L5电器基本属性信息，这三维信息以所有电器为大类进行统一编号，数值为对应编号值；L6电器类型为反应电器工作原理的外在属性特征，隶属于电器名称，因此编号为电器编号与类型编号的组合，数值为两个编号并列组合；同样L7功能模式属性类似；L 8为案例场景中的运行功率值；
L9- L12外在气象因素、日历因素，时间因素，季节因素为案例辅加信息，以辅助设备辨识。L13- L14为电器运行的电压电流时间序列值，案例的主体部分，非侵入终端辨识的主要依据，L15为辨识结果序列值，用于非侵入终端辨识的结果对比分析。
Level2- level4的案例模型如表2所示，Level2- level4为两台或多台电器设备叠加运行的场景模式，案例中涉及多个电器的特征属性为多元素组合成的数值向量。 Level2- level4的案例模型包含与Leve1相同的15个数值型的属性变量，其中属性L3- L8为多维数特征向量，维数即为叠加的电器工况数，level2的案例是2维，level3的案例是3维；其他属性信息则与level1的相应的属性信息表述一致。
案例录波方法包括如下步骤
1）选取录制地点：选取装有侵入和非侵入量测系统的试点示范家庭、对比验证平台为场所进行录制；
2）对于level1案例，按照其 L 1- L 8设定运行场景，启动对应的电器运行，录制关口处的电压及电流，作为 L 13 - L 14 序列值，同时进行分项电器的电流、电压监测，统计出电器名称、电量、和开断时间，需要辨识的结果，作为 L 15 序列值；
3）对于level2- level4案例，按照 L 3 - L 8 设定运行场景，根据采集的电器的运行情况中的错开的启动时间，依次启动及关闭对应的电器，录制关口处的电压及电流，作为 L 13 - L 14 序列值，同时进行分项电器的电流、电压监测，统计出各分项电器名称、电量、开断时间等，需要辨识的结果，作为 L 15 序列值。
案例入库方法设计，包括如下步骤
1）提炼案例模型的属性信息构建案例模型表结构；
2）根据案例模型表结构构建对应的EXCEL文档结构；
3）将步骤5得到的案例波形及属性信息进行EXCEL中录入，批量化导入系统数据库；
4）根据案例模型表结构开发案例互动化前台界面，实现案例的增、删、改、查。
非侵入终端模拟检测平台，其特征在于包括测试主机、串口服务器、非侵入辨识终端和大电流波形回放仪，测试主机通过串口服务器和非侵入辨识终端连接，大电流波形回放仪设置有以太网接口、电压输出端口和电流输出端口，以太网接口与测试主机实现信号互联，电压输出端口通过电压互感器与非侵入辨识终端连接，电流输出端口通过电流互感器与非侵入辨识终端连接。测试主机接收案例库，测试主机接收案例库信息，以及辨识非侵入辨识终端的结果，并生成报告。测试主机输出测试案例的电压电流波形（L13- L14）至大电流波形回放仪，波形回放仪复现模拟真实用户的端口电压电流，非侵入终端通过电流互感器及电压接口采集电压、电流，并启动辨识模块进行负荷辨识，将辨识结果通过串口服务器上送至测试主机，测试主机将辨识结果与真实结果（L15）进行比对评价。
非侵入辨识终端设置有485串口，485串口与串口服务器连接；电压互感器的工作电压为220V，电流互感器的工作电流为100A。非侵入辨识终端有若干个。
本发明方案所公开的技术手段不仅限于上述技术手段所公开的技术手段，还包括由以上技术特征任意组合所组成的技术方案。
以上述依据本发明的理想实施例为启示，通过上述的说明内容，相关工作人员完全可以在不偏离本项发明技术思想的范围内，进行多样的变更以及修改。本项发明的技术性范围并不局限于说明书上的内容，必须要根据权利要求范围来确定其技术性范围。
 
 
表1 level1案例模型
Figure dest_path_image002
 
表2 level2-level4案例模型
Figure dest_path_image004
 
表3常见中大功率家用电器调研信息汇总表
Figure dest_path_image006
 
表4家电运行情况调研信息汇总表
Figure dest_path_image008
 

Claims (10)

  1. 一种非侵入终端辨识能力的测试案例库构建方法，其特征在于包括以下操作步骤：
    步骤1：基础调研：选定建立数据库的区域，在该区域内进行数据采集，采集家用电器的基本信息以及各家用电器的运行情况，并做记录；
    步骤2：案例设计：根据步骤一采集的家用电器的基本信息建立案例的模型；根据步骤一采集的家用电器的运行情况建立案例库框架，进行分层分组等级化模型设计；
    步骤3：案例库建设：对案例进行案例录波和案例入库完成案例库的构建。
  2. 根据权利要求1所述的非侵入终端辨识能力的测试案例库构建方法，其特征在于所述基本信息包括电器大类、电器名称、电器品牌和电器运行原理，电器运行原理包括电器类型、电器功率型号和电器功能模式，基本信息调研方法包括入户调研、线上电商销售数据调研、线下实体店销售数据调研和利用已有的非侵入量测示范方程的数据；所述各电器的运行情况包括电器叠加使用规律、电器使用的时间规律和电器叠加运行的开启时间间隔规律，电器运行情况的调研手段包括入户调研和利用已有的非侵入量测示范方程的数据。
  3. 根据权利要求1所述的非侵入终端辨识能力的测试案例库构建方法，其特征在于所述案例库框架包括level1、level2、level3和level4四组不同等级复杂程度的运行场景，所述level1组中包括若干个案例，每个案例为单台电器单模式运行，叠加背景电器；所述level2组中包括若干个案例，每个案例为两台电器两种模式，错开启动时间，叠加运行，叠加背景电器；所述level3包括若干个案例，每个案例为三台电器三种模式，错开启动时间，叠加运行，叠加背景电器；所述level4组中包括若干个案例，每个案例为四台电器四种模式，错开启动时间，叠加运行，叠加背景电器；所述叠加方式均符合采集的电器运行情况。
  4. 根据权利要求3所述的非侵入终端辨识能力的测试案例库构建方法，其特征在于所述level1的案例模型包括15个数值型的属性变量；所述15个属性变量中，L1- L2为案例的级别属性和总运行时长，L3- L5为电器大类、电器名称和电器品牌，所述三维信息以所有电器进行统一编号，L3- L5的数值为对应编号值；L6为电器原理类型，编号为电器编号与类型编号的组合，数值为两个编号并列组合；L7为功能运行模式，编号为电器编号与功能编号的组合，数值为两个编号并列组合；L 8为运行功率值；L9- L12为运行时刻、运行日历、季节和天气温度，分别对应记录时间、日期、季节编号和温度；L13- L14为负荷电流和负荷电压，分别用电流时间序列值和电压时间序列值表示；L15为辨识结果序列值，用于非侵入终端辨识的结果对比分析；
    所述Level2- level4的案例模型包含与Leve1相同的15个数值型的属性变量，其中属性L3- L8为多维数特征向量，维数为叠加的电器工况数，所述level2的案例是2维，level3的案例是3维；其他属性信息则与level1的相应的属性信息表述一致。
  5. 根据权利要求1所述的非侵入终端辨识能力的测试案例库构建方法，其特征在于所述案例录波包括如下步骤
    1）选取录制地点：选取装有侵入和非侵入量测系统的试点示范家庭、对比验证平台为场所进行录制；
    2）对于level1案例，按照其 L 1- L 8设定运行场景，启动对应的电器运行，录制关口处的电压及电流，作为 L 13 - L 14 序列值，同时进行分项电器的电流、电压监测，统计出电器名称、电量、和开断时间，记录分项电器的电流、电压、电器名称、电量和开断时间作为 L 15 序列值，即为需要辨识的结果；
    3）对于level2- level4案例，按照 L 3 - L 8 设定运行场景，根据采集的电器的运行情况中的错开的启动时间，依次启动及关闭对应的电器，录制关口处的电压及电流，作为 L 13 - L 14 序列值，同时进行分项电器的电流、电压监测，统计出各分项电器名称、电量和开断时间，需要辨识的结果。
  6. 根据权利要求5所述的非侵入终端辨识能力的测试案例库构建方法，其特征在于所述案例入库，包括如下步骤
    1）根据案例模型的属性信息构建案例模型表结构；
    2）根据案例模型表结构构建对应的EXCEL文档结构；
    3）将步骤5得到的案例波形及属性信息录入EXCEL中，批量化导入系统数据库获得测试案例库；
    4）根据案例模型表结构开发案例互动化前台界面，实现案例的增、删、改、查。
  7. 非侵入终端模拟检测平台，其特征在于包括测试主机、串口服务器、非侵入辨识终端和大电流波形回放仪，所述测试主机通过串口服务器和非侵入辨识终端连接，所述大电流波形回放仪设置有以太网接口、电压输出端口和电流输出端口，所述以太网接口与测试主机实现信号互联，所述电压输出端口通过电压互感器与非侵入辨识终端连接，所述电流输出端口通过电流互感器与非侵入辨识终端连接。
  8. 根据权利要求7所述的非侵入终端模拟检测平台，其特征在于所述非侵入辨识终端设置有485串口，所述485串口与串口服务器连接；所述电压互感器的工作电压为220V，电流互感器的工作电流为100A。
  9. 根据权利要求7所述的非侵入终端模拟检测平台，其特征在于所述非侵入辨识终端有若干个。
  10. 根据权利要求1所述的非侵入终端模拟检测平台，其特征在于所述测试主机接收案例库信息，以及辨识非侵入辨识终端的结果，并生成报告；测试主机输出测试案例的电压电流波形，即 L 13 - L 14 至大电流波形回放仪，波形回放仪复现模拟真实用户的端口电压电流，非侵入终端通过电流互感器及电压接口采集电压、电流，并启动辨识模块进行负荷辨识，将辨识结果通过串口服务器上送至测试主机，测试主机将辨识结果与真实结果即 L 15 进行比对评价。
PCT/CN2019/113106 2019-04-24 2019-10-24 非侵入终端辨识能力测试案例库构建方法及模拟检测平台 WO2020215636A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910335046.1 2019-04-24
CN201910335046.1A CN110033395B (zh) 2019-04-24 2019-04-24 非侵入终端辨识能力测试案例库构建方法及模拟检测平台

Publications (1)

Publication Number Publication Date
WO2020215636A1 true WO2020215636A1 (zh) 2020-10-29

Family

ID=67240054

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/113106 WO2020215636A1 (zh) 2019-04-24 2019-10-24 非侵入终端辨识能力测试案例库构建方法及模拟检测平台

Country Status (2)

Country Link
CN (1) CN110033395B (zh)
WO (1) WO2020215636A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110033395B (zh) * 2019-04-24 2023-08-01 江苏智臻能源科技有限公司 非侵入终端辨识能力测试案例库构建方法及模拟检测平台
CN110944413B (zh) * 2019-12-06 2021-08-10 江苏智臻能源科技有限公司 云边协同架构下基于历史负荷辨识数据的电热细分方法
CN111007450A (zh) * 2019-12-06 2020-04-14 江苏智臻能源科技有限公司 一种检测负荷辨识设备的结果可信度的方法
CN110995543B (zh) * 2019-12-18 2022-11-01 云南大学 一种非侵入式未成年人异常上网行为监测方法
CN111562434A (zh) * 2020-04-28 2020-08-21 国电南瑞科技股份有限公司 一种非入户电器信息智能量测系统及方法
CN111932406B (zh) * 2020-09-18 2021-01-08 江苏智臻能源科技有限公司 一种叠加运行工况下负荷辨识效果的评价方法
CN112014788B (zh) * 2020-11-02 2021-02-02 江苏智臻能源科技有限公司 基于录波文件回放的负荷辨识模组检测方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107730003A (zh) * 2017-10-23 2018-02-23 华中科技大学 一种支持多电器类型高精度的nilm实现方法
US10120004B1 (en) * 2017-10-20 2018-11-06 Institute For Information Industry Power consumption analyzing server and power consumption analyzing method thereof
CN110033395A (zh) * 2019-04-24 2019-07-19 江苏智臻能源科技有限公司 非侵入终端辨识能力测试案例库构建方法及模拟检测平台

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106646026A (zh) * 2016-11-11 2017-05-10 华北电力大学 一种非侵入式家电负荷识别方法
CN108762502A (zh) * 2018-05-24 2018-11-06 山东师范大学 一种基于眼动追踪的虚拟现实人群仿真方法及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10120004B1 (en) * 2017-10-20 2018-11-06 Institute For Information Industry Power consumption analyzing server and power consumption analyzing method thereof
CN107730003A (zh) * 2017-10-23 2018-02-23 华中科技大学 一种支持多电器类型高精度的nilm实现方法
CN110033395A (zh) * 2019-04-24 2019-07-19 江苏智臻能源科技有限公司 非侵入终端辨识能力测试案例库构建方法及模拟检测平台

Also Published As

Publication number Publication date
CN110033395B (zh) 2023-08-01
CN110033395A (zh) 2019-07-19

Similar Documents

Publication Publication Date Title
WO2020215636A1 (zh) 非侵入终端辨识能力测试案例库构建方法及模拟检测平台
CN106096726B (zh) 一种非侵入式负荷监测方法及装置
Pereira et al. Sustdata: A public dataset for ict4s electric energy research
Yu et al. A methodology for identifying and improving occupant behavior in residential buildings
CN110097297A (zh) 一种多维度窃电态势智能感知方法、系统、设备及介质
CN109325545A (zh) 低压配电网拓扑结构校验方法、装置、设备及存储介质
CN106570784A (zh) 电压监测一体化模型
CN106646026A (zh) 一种非侵入式家电负荷识别方法
CN106327979A (zh) 用电信息采集模拟仿真培训系统
WO2021109517A1 (zh) 一种检测负荷辨识设备的结果可信度的方法
JP3788555B2 (ja) 負荷種類分析方法及び装置、消費性向診断方法、装置及びシステム並びに記録媒体
CN108594041A (zh) 一种针对非侵入式家居电力负荷监测装置的检测平台
CN109815994A (zh) 一种用户用电负荷特征分析方法
CN108054749A (zh) 一种非侵入式电力负荷分解方法及装置
CN106990774A (zh) 一种电动汽车便携式故障诊断仪及其监测方法
CN107219492B (zh) 中压电网电能计量高压一体化半物理仿真试验装置
CN104392603B (zh) 一种现场用电信息采集设备智能诊断系统
CN109633517A (zh) 分相分时查找窃电的方法
Guohua et al. Research on non-intrusive load monitoring based on random forest algorithm
CN206774638U (zh) 一种大型储能系统
Ramadan et al. Simulating appliance-based power consumption records for energy efficiency awareness
CN206628080U (zh) 基于业务系统采维一体化培训及考核装置
CN113762911A (zh) 一种基于分布式架构的综合能源服务平台的制作方法
Yan et al. Comprehensive energy efficiency rating evaluation model of enterprise power based on grid big data
Hashim et al. Smart building energy management using big data analytic approach

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: 19926244

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19926244

Country of ref document: EP

Kind code of ref document: A1