WO2024098990A1 - 一种基于专变采集终端的电能质量监测方法及系统 - Google Patents

一种基于专变采集终端的电能质量监测方法及系统 Download PDF

Info

Publication number
WO2024098990A1
WO2024098990A1 PCT/CN2023/121728 CN2023121728W WO2024098990A1 WO 2024098990 A1 WO2024098990 A1 WO 2024098990A1 CN 2023121728 W CN2023121728 W CN 2023121728W WO 2024098990 A1 WO2024098990 A1 WO 2024098990A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
abnormal
source
power quality
acquisition
Prior art date
Application number
PCT/CN2023/121728
Other languages
English (en)
French (fr)
Inventor
倪志伟
汪升川
高平航
丁剑飞
侯国平
Original Assignee
浙江万胜智能科技股份有限公司
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 浙江万胜智能科技股份有限公司 filed Critical 浙江万胜智能科技股份有限公司
Publication of WO2024098990A1 publication Critical patent/WO2024098990A1/zh

Links

Classifications

    • 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

Definitions

  • the present invention relates to the field of data processing technology, and in particular to a method and system for monitoring power quality based on a dedicated transformer acquisition terminal.
  • the power quality reflects whether the power grid and power system are in a safe operating state. Good power quality is an important guarantee for the normal operation of economic production and the quality of people's lives. Accurate and real-time grasp of the power quality status will help to ensure the normal operation of various types of electrical equipment and avoid reversible or irreversible damage to electrical equipment.
  • the present application provides a power quality monitoring method and system based on a dedicated transformer collection terminal, which is used to solve the technical problem that in the prior art, during the power quality assessment process for dedicated transformer users, the power quality assessment data collection accuracy is insufficient, resulting in low reference value of the power quality assessment results in power operation and maintenance.
  • the present application provides a method and system for monitoring power quality based on a dedicated transformer acquisition terminal.
  • a method for monitoring power quality based on a dedicated transformer acquisition terminal comprising: connecting the power quality monitoring system to obtain a monitoring indicator set for power quality assessment; determining a data acquisition source for the dedicated transformer acquisition terminal based on the monitoring indicator set; determining a plurality of data transmission channels using the data acquisition source as a connection object; obtaining an abnormal feature set by performing data feature recognition on the data acquisition source, wherein the abnormal feature set is a data set containing abnormalities in real-time acquired data; performing model training using the abnormal feature set as a training set to obtain a plurality of abnormal recognition models, wherein the plurality of abnormal recognition models are respectively embedded in the plurality of data transmission channels; outputting an abnormal recognition result based on the plurality of abnormal recognition models; inputting the abnormal recognition result into the data conversion module, and obtaining a data conversion result based on the data conversion module; transmitting the data conversion result to the power quality monitoring system via the plurality of data transmission channels for power quality assessment.
  • a power quality monitoring system based on a special transformer acquisition terminal, the system comprising: a detection index acquisition module, used to connect to the power quality monitoring system and obtain a monitoring index set for power quality assessment; a data source determination module, used to determine the data acquisition source for the special transformer acquisition terminal according to the monitoring index set; a data channel determination module, used to determine multiple data transmission channels with the data acquisition source as the connection object; a data feature recognition module, used to obtain an abnormal feature set by performing data feature recognition on the data acquisition source, wherein the abnormal feature set is a data set with abnormalities in real-time acquisition data; a recognition model training module, used to perform model training with the abnormal feature set as a training set, and obtain multiple abnormal recognition models, wherein the multiple abnormal recognition models are respectively embedded in the multiple data transmission channels; an abnormal recognition output module, used to output abnormal recognition results according to the multiple abnormal recognition models; a data conversion execution module, used to input the abnormal recognition results into a data conversion module, and obtain data conversion results according to
  • the method provided in the embodiment of the present application obtains a monitoring indicator set for power quality assessment by connecting to the power quality monitoring system; determines a data acquisition source for a special transformer acquisition terminal according to the monitoring indicator set; provides a reference basis for the subsequent construction of a data transmission channel, determines multiple data transmission channels with the data acquisition source as a connection object; obtains an abnormal feature set by performing data feature recognition on the data acquisition source, wherein the abnormal feature set is a data set with abnormalities in real-time collected data; performs model training with the abnormal feature set as a training set to obtain multiple abnormal recognition models, wherein the multiple abnormal recognition models are respectively embedded in the multiple data transmission channels, so as to facilitate abnormal recognition of multiple types of data in units of data channels, improve the accuracy and efficiency of abnormal data recognition, and output abnormal recognition results according to the multiple abnormal recognition models; inputs the abnormal recognition result into the data conversion module, obtains the data conversion result according to the data conversion module, realizes the correction of abnormal data to normal data, and improves the validity of reference data for power quality assessment in the power quality monitoring model
  • FIG1 is a schematic diagram of a flow chart of a method for monitoring power quality based on a dedicated transformer acquisition terminal provided by the present application;
  • FIG2 is a schematic diagram of a flow chart of identifying and reminding abnormal collection devices in a power quality monitoring method based on a dedicated transformer collection terminal provided by the present application;
  • FIG3 is a schematic diagram of a flow chart of obtaining data conversion results in a power quality monitoring method based on a dedicated transformer acquisition terminal provided by the present application;
  • FIG4 is a schematic diagram of the structure of a power quality monitoring system based on a dedicated transformer acquisition terminal provided in the present application.
  • detection index acquisition module 11 data source determination module 12, data channel determination module 13, data feature recognition module 14, recognition model training module 15, abnormality recognition output module 16, data conversion execution module 17, power quality assessment module 18.
  • the present application provides a method and system for monitoring power quality based on a dedicated transformer acquisition terminal, which is used to solve the technical problem that in the prior art, during the power quality assessment process for dedicated transformer users, the power quality assessment data acquisition accuracy is insufficient, resulting in a low reference value of the power quality assessment results in power operation and maintenance.
  • the method achieves the technical effect of improving the data acquisition accuracy related to power quality assessment for dedicated transformer users, improving the referenceability and effectiveness of power quality assessment results for dedicated transformer users in power operation and maintenance, and ensuring the power safety of dedicated transformer users.
  • the present application provides a method for monitoring power quality based on a dedicated transformer acquisition terminal, the method being applied to a power quality monitoring system, the system being communicatively connected to a data conversion module, the method comprising:
  • the power quality is a dynamic evaluation result of the quality of power in the power system obtained by comprehensive evaluation based on multiple data indicators such as voltage deviation, frequency deviation, harmonics and simple harmonics.
  • Monitoring the power quality helps to analyze and determine whether there is a problem with the current power quality, so as to take control strategies in time to eliminate or suppress power dangers, thereby ensuring the safe operation of various types of electrical equipment.
  • the power quality is dynamically monitored in real time based on the power quality monitoring system, and the power quality monitoring system outputs the power quality evaluation result by acquiring multiple power data monitoring indicators of the power user terminal.
  • the power quality monitoring system based on the dedicated transformer acquisition terminal is communicatively connected to the power quality monitoring system, and the power quality monitoring system based on the dedicated transformer acquisition terminal determines the monitoring data used to evaluate the power quality according to the power quality monitoring system, and obtains the monitoring indicator set used for evaluating the power quality of the dedicated transformer user in the power quality monitoring system based on the dedicated transformer acquisition terminal.
  • S200 Determine a data collection source for a dedicated transformer collection terminal according to the monitoring indicator set;
  • dedicated transformer users are users who use their own property transformers to supply electricity, including large industrial dedicated transformers, non-general industrial dedicated transformers, commercial mixed electricity users, etc.
  • dedicated transformer users Compared with public transformer users, dedicated transformer users generally lack professional electricians and have knowledge blind spots in electricity business. Therefore, during the electricity use process, they cannot promptly know whether there are dangerous conditions such as overload operation, which causes electrical equipment to be affected by power fluctuations.
  • the dedicated transformer acquisition terminal is used to collect a variety of data for power quality assessment in real time from a variety of power data of dedicated transformer users.
  • the data acquisition source is a data acquisition device that monitors the dynamic power data of the power system.
  • multiple types of monitoring data for evaluating the power quality of special transformer users are determined, and based on the multiple types of monitoring data, multiple data collection sources for the special transformer collection terminal to collect data for the special transformer users are determined.
  • S400 Acquire an abnormal feature set by performing data feature recognition on the data collection source, wherein the abnormal feature set is a data set with abnormalities in the real-time collected data;
  • the data transmission channel is used to transmit the data obtained by the data acquisition device at the data acquisition source to the power quality monitoring system for power quality evaluation.
  • the data transmission channels have a corresponding relationship with the data acquisition sources, and each data transmission channel transmits data of the same data acquisition source.
  • Defective data existing in historical data collection of multiple data collection sources are obtained, multiple data defect features corresponding to the multiple data collection sources are obtained based on the historical defective data of the data collection sources, and the real-time output data of the corresponding multiple data collection sources are traversed based on the multiple data defect features to perform data feature recognition, and the abnormal feature set existing in the real-time collection data is obtained, the abnormal feature set reflects the data defects currently existing in the real-time collection data obtained based on the data collection sources, and the abnormal feature set is composed of abnormal data from multiple data collection sources.
  • S500 Performing model training using the abnormal feature set as a training set to obtain multiple abnormality recognition models, wherein the multiple abnormality recognition models are respectively embedded in the multiple data transmission channels;
  • the method step S500 of acquiring multiple anomaly recognition models further includes:
  • multiple anomaly recognition models are set corresponding to multiple data acquisition sources.
  • This embodiment does not impose any restrictions on the construction and training methods of data anomaly recognition models.
  • the defect data existing in the historical data acquisition of multiple data acquisition sources and the abnormal feature data sets of multiple data acquisition sources are used as training data for model training to obtain multiple anomaly recognition models whose model output results have reached the preset output accuracy requirements.
  • the multiple anomaly recognition models are embedded in the multiple data transmission channels that have a corresponding relationship with the multiple data acquisition sources.
  • the anomaly recognition model embedded in the data transmission channel is in real-time operation during the data transmission process of the data acquisition source, resulting in a waste of system computing resources and reduced data transmission efficiency.
  • activation conditions are set for the data anomaly recognition model, and a data transmission quality coefficient is obtained by performing data transmission quality evaluation on the real-time collected data in terms of data transmission delay and data loss. Then, it is determined whether to activate multiple anomaly recognition models of multiple data transmission channels based on the obtained data transmission quality coefficient.
  • the transmission quality evaluation model is constructed to evaluate the data transmission quality of the real-time collected data in terms of data transmission delay and data transmission loss.
  • the transmission quality assessment model is connected to the multiple anomaly recognition models, and the transmission quality assessment model is linked to the multiple data transmission channels.
  • the real-time acquisition data transmitted in the multiple data transmission channels are subjected to transmission quality assessment to obtain multiple transmission quality coefficients.
  • the multiple transmission quality coefficients are used as constraints for activating the multiple anomaly recognition models, and the transmission quality coefficients are compared with activation coefficients preset for the multiple anomaly recognition models to determine the corresponding anomaly recognition model that needs to be activated to perform anomaly recognition of the real-time acquisition data.
  • This embodiment embeds multiple anomaly recognition models in multiple data transmission channels accordingly, and links a transmission quality assessment model that evaluates the data transmission quality in the data transmission channel to perform activation judgments on multiple anomaly recognition models, thereby achieving the technical effect of accurately identifying abnormal data in real-time collected data while avoiding the impact of abnormal data identification on data transmission efficiency.
  • the abnormality identification result is a data deviation generated during the data transmission process of the data transmission channel.
  • the abnormality identification of the data in each data transmission channel is performed according to the multiple abnormality identification models embedded in the multiple data transmission channels to obtain the abnormality identification result, and the data of the abnormality identification result is restored and corrected, so that the relevant data planned to be transmitted in each data transmission channel to the power quality detection system for power quality assessment are credible.
  • Step S700 of the method provided in the present application further includes:
  • S710 Analyze the abnormality source of the abnormality identification result to obtain an identified abnormality source, wherein the identified abnormality source is the data source with the largest proportion of abnormal data;
  • S720 Performing model training with the identified abnormal source to generate a data conversion model, wherein the data conversion model is used to implement calibration and correction of abnormal data;
  • S730 Correct the abnormality recognition result with the data conversion module embedded in the data conversion model, and output the data conversion result.
  • the data conversion module is used to restore and correct the abnormality identification result into normal data
  • the identified abnormality source is the data source with the largest proportion in the abnormal data.
  • the cause of the data abnormality is determined, such as the data abnormality caused by signals interfering with data transmission when some devices use remote data transmission or signal transmission.
  • the data source with the largest proportion of abnormal data is obtained as the identified abnormal source, and the identified abnormal source is representative of the data abnormality.
  • the model is trained with the identified abnormal source to generate a data conversion model for calibration and correction of abnormal data.
  • This embodiment does not limit the model construction method and training method of the data conversion model, which can be set according to actual needs.
  • the trained data conversion model is embedded in the data conversion module of the data conversion model, the abnormal recognition result is corrected, and the data conversion result is output.
  • This embodiment performs abnormal source analysis on various types of abnormal data, obtains data sources with common data anomalies, and constructs and trains abnormal data correction models, thereby achieving the technical effect of performing targeted correction on abnormal data, improving the transmission quality of data transmission channels, and indirectly improving the effectiveness and credibility of power quality assessment results.
  • S800 Transmitting the data conversion results via the multiple data transmission channels to the power quality monitoring system for power quality assessment.
  • the data conversion results and the real-time collected data without data anomalies are transmitted to the power quality monitoring system through the multiple data transmission channels, and the power quality monitoring system performs power quality assessment and outputs the power quality assessment results.
  • the method provided in this embodiment obtains a monitoring indicator set for power quality assessment by connecting to the power quality monitoring system; determines a data acquisition source for a special transformer acquisition terminal according to the monitoring indicator set; provides a reference basis for the subsequent construction of a data transmission channel, determines multiple data transmission channels with the data acquisition source as a connection object; obtains an abnormal feature set by performing data feature recognition on the data acquisition source, wherein the abnormal feature set is a data set with abnormalities in real-time collected data; performs model training with the abnormal feature set as a training set to obtain multiple abnormal recognition models, wherein the multiple abnormal recognition models are respectively embedded in the multiple data transmission channels, so as to facilitate abnormal recognition of multiple types of data in units of data channels, improve the accuracy and efficiency of abnormal data recognition, and output abnormal recognition results according to the multiple abnormal recognition models; inputs the abnormal recognition result into the data conversion module, obtains the data conversion result according to the data conversion module, realizes the correction of abnormal data to normal data, and improves the validity of reference data for power quality assessment in the power quality monitoring model; transmits the
  • the method steps provided by the present application also include:
  • S210 Acquire information of the data acquisition device of the data acquisition source
  • S220 extracting equipment operating condition data from the information of the data acquisition device to obtain real-time equipment operating condition information
  • S240 If the data acquisition device is in an abnormal state, mark and remind the abnormal data acquisition device.
  • S231 Determine whether the data acquisition device is a complete set of acquisition devices, wherein the complete set of acquisition devices is a sub-device and a main device that assists in executing data acquisition;
  • the data acquisition device is a complete set of acquisition devices, and a data transmission path of the complete set of acquisition devices is obtained;
  • S233 Determine a data transmission abnormality source based on the data transmission path
  • S234 Locate the abnormal data collection device according to the abnormal data transmission source.
  • this embodiment verifies whether the current data acquisition equipment has abnormal status before performing data acquisition of the special transformer user at the data acquisition source through the data acquisition equipment.
  • information of the data acquisition device of the data acquisition source is obtained, and equipment operating condition data of the data acquisition device is obtained based on the information of the data acquisition device.
  • the equipment operating condition data and the equipment operating condition data acquisition are actually used as the real-time operating condition information of the acquired equipment.
  • an operating condition data fluctuation curve image is generated, and whether the data acquisition device is in an abnormal state is judged based on whether the changes in the data fluctuation curve are regular.
  • the device composition type of the data acquisition device is further determined.
  • the device composition types of the data acquisition device include a single data acquisition device and a complete set of acquisition devices.
  • a single data acquisition device is a device that performs data source data acquisition based on a single device.
  • a complete set of data acquisition equipment is a combination of devices that perform data acquisition by cooperating with a parent device including a sub-device that assists in executing data acquisition.
  • the data transmission path in the complete set of data acquisition equipment is from the sub-device to the parent device, and there is a one-to-one correspondence or a one-to-many correspondence between the parent device and the sub-device.
  • an abnormal device identification reminder is issued to the data acquisition device, and the data acquisition operation of the data acquisition device is stopped.
  • the data acquisition device is a complete set of acquisition equipment
  • the sub-devices and the main device that assist in executing data acquisition in the complete set of acquisition equipment are further determined, the sub-devices and the main device that specifically cooperate in data transmission in the complete set of acquisition equipment are obtained, and the data transmission path is obtained.
  • the data transmission path it is determined whether the source of the data transmission anomaly is in the data transmission process or at the data transmission source.
  • the abnormal acquisition device is located.
  • the abnormal acquisition main device is located.
  • the abnormal acquisition sub-device is located accordingly. The abnormal acquisition device is marked and reminded for data screening during subsequent data source data acquisition to avoid erroneous data from flowing into the power quality monitoring system, which reduces the reference value of the power quality assessment results.
  • This embodiment analyzes the operating conditions of the equipment used for data collection before executing data source data collection, identifies data collection equipment with abnormal conditions, and avoids abnormal data from flowing into the power quality detection system to participate in power quality assessment, thereby achieving the technical effect of improving the effectiveness and reference of power quality assessment results for special transformer users.
  • step S520 of the method provided by the present application further includes:
  • S521 Determine a plurality of groups of test samples according to the plurality of data transmission channels, wherein the plurality of groups of test samples correspond to the plurality of data transmission channels, and each group of test samples includes a plurality of data sets;
  • S522 Performing a transmission frame period test on the multiple data transmission channels based on the multiple test samples, and outputting multiple transmission delay indexes, wherein the multiple transmission delay indexes correspond one-to-one to the multiple data transmission channels;
  • step S523 of the method provided by the present application further includes:
  • S523-1 Based on the multiple test samples, obtain multiple output samples obtained through the multiple data transmission channels;
  • S523-2 Perform data loss analysis on the multiple test samples and the multiple output samples, and output multiple data loss indexes, wherein the multiple data loss indexes correspond to the multiple data transmission channels one by one;
  • S523-3 Calculate the multiple data loss indexes as a second data set to generate the multiple transmission quality coefficients.
  • the transmission frame period test generates a transmission frame period by performing a transmission period test based on the response time from the time when the transmission data is sent from the data acquisition device as the sending end to the time when the transmission data arrives at the power quality monitoring system as the receiving end via the data transmission channel to receive the transmission data.
  • multiple groups of test samples are determined according to the multiple data transmission channels, the data types of the multiple groups of test samples correspond to the transmission data types of the multiple data transmission channels, and each group of test samples includes multiple data sets with different data sizes.
  • a transmission frame period test is performed on the multiple data transmission channels based on the multiple test samples, and multiple transmission delay indexes are output according to the data transmission frame periods of the multiple test samples in the multiple data transmission channels, where the transmission delay indexes are in units of time.
  • multiple output samples obtained through the multiple data transmission channels are obtained, and data loss analysis is performed based on the multiple test samples and the multiple output samples.
  • Multiple data loss indexes are output based on the degree of data loss between the test samples and the output samples.
  • the data loss index is in units of data occupied space, and the multiple data loss indexes correspond one-to-one to the multiple data transmission channels.
  • the multiple transmission delay indices and the multiple data loss indices are normalized, and the multiple transmission delay indices are used as the first data set and the multiple data loss indices are used as the second data set for calculation to generate the multiple transmission quality coefficients, and the multiple transmission quality coefficients correspond one-to-one to the multiple data transmission channels.
  • This embodiment constructs a transmission quality assessment model and links it into multiple data transmission channels to evaluate and determine whether it is necessary to activate the abnormality recognition model embedded in each data transmission channel in terms of data transmission delay and data transmission loss. This achieves the technical effect of reducing the consumption of system computing resources by the abnormality recognition model, and avoiding the need to perform abnormal data recognition analysis on all data in the data transmission channel, which causes a decrease in data transmission efficiency and reduces the timeliness of the power quality obtained by the power quality monitoring model.
  • the present application provides a power quality monitoring system based on the dedicated transformer acquisition terminal, wherein the system comprises:
  • the detection index acquisition module 11 is used to connect to the power quality monitoring system and obtain a monitoring index set for power quality assessment;
  • a data source determination module 12 for determining a data collection source for a dedicated transformer collection terminal according to the monitoring indicator set;
  • a data feature recognition module 14 is used to obtain an abnormal feature set by performing data feature recognition on the data collection source, wherein the abnormal feature set is a data set with abnormalities in the real-time collected data;
  • a recognition model training module 15 is used to perform model training using the abnormal feature set as a training set to obtain multiple abnormal recognition models, wherein the multiple abnormal recognition models are respectively embedded in the multiple data transmission channels;
  • An abnormality identification output module 16 used for outputting abnormality identification results according to the multiple abnormality identification models
  • the power quality assessment module 18 is used to transmit the data conversion results through the multiple data transmission channels to the power quality monitoring system for power quality assessment.
  • the data source determination module 12 also includes:
  • a device information obtaining unit used to obtain information about the data acquisition device of the data acquisition source
  • An equipment working condition acquisition unit used to extract equipment working condition data from the information of the data acquisition device to obtain real-time equipment working condition information
  • An equipment status judgment unit used to judge whether the data acquisition equipment is in an abnormal state according to the real-time working condition information of the equipment
  • the device abnormality identification unit is used to identify and remind the abnormal data collection device if the data collection device is in an abnormal state.
  • the device status determination unit further includes:
  • the device constitutes a judgment unit, which is used to judge whether the data acquisition device is a complete set of acquisition devices, wherein the complete set of acquisition devices is a sub-device and a main device that assists in executing data acquisition;
  • a data transmission judgment unit used for obtaining the data transmission path of the complete set of data collection equipment if the data collection equipment is a complete set of data collection equipment;
  • An abnormality source determination unit used to determine a data transmission abnormality source based on the data transmission path
  • the abnormal device locating unit is used to locate the abnormal data collection device according to the abnormal data transmission source.
  • the recognition model training module 15 also includes:
  • An evaluation model building unit used to build a transmission quality evaluation model, wherein the transmission quality evaluation model is connected to the multiple anomaly recognition models;
  • An evaluation model linking unit used to link the transmission quality evaluation model to the multiple data transmission channels, perform transmission quality evaluation on the multiple data transmission channels, and obtain multiple transmission quality coefficients;
  • the constraint condition preset unit is used to use the multiple transmission quality coefficients as constraint conditions for activating the multiple abnormality recognition models.
  • the data channel determination module 13 also includes:
  • a test sample determination unit configured to determine a plurality of groups of test samples according to the plurality of data transmission channels, wherein the plurality of groups of test samples correspond to the plurality of data transmission channels, and each group of test samples includes a plurality of data sets;
  • a transmission period test unit configured to perform a transmission frame period test on the plurality of data transmission channels based on the plurality of test samples, and output a plurality of transmission delay indexes, wherein the plurality of transmission delay indexes correspond one-to-one to the plurality of data transmission channels;
  • the transmission quality calculation unit is used to calculate the multiple transmission delay indexes as the first data set to generate the multiple transmission quality coefficients.
  • the transmission quality calculation unit further includes:
  • An output sample obtaining unit configured to obtain, based on the multiple test samples, multiple output samples obtained through the multiple data transmission channels
  • a data loss analysis unit configured to perform data loss analysis on the plurality of test samples and the plurality of output samples, and output a plurality of data loss indexes, wherein the plurality of data loss indexes correspond one-to-one to the plurality of data transmission channels;
  • the transmission quality generating unit is used to calculate the multiple data loss indexes as the second data set to generate the multiple transmission quality coefficients.
  • the data conversion execution module 17 also includes:
  • An abnormal source analysis unit configured to obtain an identified abnormal source by performing abnormal source analysis on the abnormal identification result, wherein the identified abnormal source is a data source with the largest proportion of abnormal data;
  • An abnormal data correction unit used to perform model training with the identified abnormal source to generate a data conversion model, wherein the data conversion model is used to achieve calibration correction of the abnormal data;
  • a data conversion output unit is used to correct the abnormality recognition result by using the data conversion module embedded in the data conversion model and output the data conversion result.
  • Any of the methods or steps described above may be stored as computer instructions or programs in various types of computer memories, and the computer instructions or programs may be recognized by various types of computer processors to implement any of the methods or steps described above.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)

Abstract

一种基于专变采集终端的电能质量监测方法及系统，涉及数据处理技术领域，方法包括：根据电能质量评估的监测指标确定专变采集终端的数据采集源并对应构建数据传输通道，对数据采集源进行数据特征识别，输出异常识别结果，将异常识别结果进行数据纠正获得数据转换结果，将数据转换结果传输至电能质量监测系统进行电能质量评估。解决了现有技术中对专变用户进行电能质量评估过程中，存在电能质量评估数据采集精度不足，导致电能质量评估结果在电力运维中参考价值较低的技术问题。

Description

一种基于专变采集终端的电能质量监测方法及系统 技术领域
本发明涉及数据处理技术领域，具体涉及一种基于专变采集终端的电能质量监测方法及系统。
 背景技术
电能质量反映了电网和电力系统是否处于安全运行状态，良好的电能质量是经济生产正常运行和国民生活质量的重要保障，对于电能质量状况的准确实时把握，有助于保障各类用电设备的正常运行以及避免用电设备发生可逆或不可逆性损害。
随着电力科学和我国电网系统的不断发展，我国对于电能质量监测的重视程度不断加深，但现阶段进行电能质量评估的中心在于电能质量评估参考因素的科学性，而非关注所获得的电能质量评估参考因素的数据准确性。
现有技术中对专变用户进行电能质量评估过程中，存在电能质量评估数据采集精度不足，导致电能质量评估结果在电力运维中参考价值较低的技术问题。
 发明内容
本申请提供了一种基于专变采集终端的电能质量监测方法及系统，用于针对解决现有技术中对专变用户进行电能质量评估过程中，存在电能质量评估数据采集精度不足，导致电能质量评估结果在电力运维中参考价值较低的技术问题。
鉴于上述问题，本申请提供了一种基于专变采集终端的电能质量监测方法及系统。
本申请的第一个方面，提供了一种基于专变采集终端的电能质量监测方法，所述方法包括：连接所述电能质量监测系统，获取用于进行电能质量评估的监测指标集；根据所述监测指标集，确定用于专变采集终端的数据采集源；以所述数据采集源作为连接对象，确定多个数据传输通道；通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道；根据所述多个异常识别模型，输出异常识别结果；将所述异常识别结果输入所述数据转换模块中，根据所述数据转换模块，得到数据转换结果；将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。
本申请的第二个方面，提供了一种基于专变采集终端的电能质量监测系统，所述系统包括：检测指标获得模块，用于连接电能质量监测系统，获取用于进行电能质量评估的监测指标集；数据来源确定模块，用于根据所述监测指标集，确定用于专变采集终端的数据采集源；数据通道确定模块，用于以所述数据采集源作为连接对象，确定多个数据传输通道；数据特征识别模块，用于通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；识别模型训练模块，用于以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道；异常识别输出模块，用于根据所述多个异常识别模型，输出异常识别结果；数据转换执行模块，用于将所述异常识别结果输入数据转换模块中，根据所述数据转换模块，得到数据转换结果；电能质量评估模块，用于将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。
本申请中提供的一个或多个技术方案，至少具有如下技术效果或优点：
本申请实施例提供的方法通过连接所述电能质量监测系统，获取用于进行电能质量评估的监测指标集；根据所述监测指标集，确定用于专变采集终端的数据采集源；为后续进行数据传输通道的构建提供参考基础，以所述数据采集源作为连接对象，确定多个数据传输通道；通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道，便于以数据通道为单位，对多类型数据进行异常识别，提高异常数据识别准确度和识别效率，根据所述多个异常识别模型，输出异常识别结果；将所述异常识别结果输入所述数据转换模块中，根据所述数据转换模块，得到数据转换结果，实现将异常数据纠正为正常数据，提高最终在电能质量监测模型中进行电能质量评估的参考数据的有效性；将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。达到了提高进行专变用户电能质量评估相关数据采集精度，提高专变用户电能质量评估结果在电力运维中的可参考性和有效性，实现确保专变用户的用电安全的技术效果。
 附图说明
图1为本申请提供的一种基于专变采集终端的电能质量监测方法流程示意图；
图2为本申请提供的一种基于专变采集终端的电能质量监测方法中标识提醒异常采集设备的流程示意图；
图3为本申请提供的一种基于专变采集终端的电能质量监测方法中获取数据转换结果的流程示意图；
图4为本申请提供的一种基于专变采集终端的电能质量监测系统的结构示意图。
附图标记说明：检测指标获得模块11，数据来源确定模块12，数据通道确定模块13，数据特征识别模块14，识别模型训练模块15，异常识别输出模块16，数据转换执行模块17，电能质量评估模块18。
 本发明的实施方式
本申请提供了一种基于专变采集终端的电能质量监测方法及系统，用于针对解决现有技术中对专变用户进行电能质量评估过程中，存在电能质量评估数据采集精度不足，导致电能质量评估结果在电力运维中参考价值较低的技术问题。达到了提高进行专变用户电能质量评估相关数据采集精度，提高专变用户电能质量评估结果在电力运维中的可参考性和有效性，实现确保专变用户的用电安全的技术效果。
本发明技术方案中对数据的获取、存储、使用、处理等均符合国家法律法规的相关规定。
下面，将参考附图对本发明中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅是本发明的一部分实施例，而不是本发明的全部实施例，应理解，本发明不受这里描述的示例实施例的限制。基于本发明的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。另外还需要说明的是，为了便于描述，附图中仅示出了与本发明相关的部分而非全部。
实施例一
如图1所示，本申请提供了一种基于专变采集终端的电能质量监测方法，所述方法应用于电能质量监测系统，所述系统与数据转换模块通信连接，所述方法包括：
S100：连接所述电能质量监测系统，获取用于进行电能质量评估的监测指标集；
具体而言，应理解的，所述电能质量为根据电压偏差、频率偏差、谐波和简谐波等多种数据指标进行综合评估，获得的反应电力系统中电能的质量的动态评价结果。通过对电能质量进行监测有助于分析确定当前电能质量是否存在问题，以便及时采取控制策略进行电能危险问题的消除或抑制，从而确保各类型用电设备安全运行。
在本实施例中，电能质量基于所述电能质量监测系统进行实时动态监测，所述电能质量监测系统通过获取用电终端的多项电力数据监测指标输出电能质量评估结果。为实现对于专变用户的电能质量状况进行准确监测评估，将所述基于专变采集终端的电能质量监测系统与所述电能质量监测系统通信连接，所述基于专变采集终端的电能质量监测系统根据所述电能质量监测系统确定用于评估电能质量的监测数据，获得在所述基于专变采集终端的电能质量监测系统中用于进行专变用户电能质量评估的所述监测指标集。
S200：根据所述监测指标集，确定用于专变采集终端的数据采集源；
具体而言，专变用户为使用自身产权变压器供电的用户，包括大工业专变、非普工业专变、商业混合用电等专变用户，由于专变用户相较于公变用户普遍存在缺少专职电工，在电力业务方面存在知识盲区，因而在用电过程中无法及时获知当前是否存在过负荷运行等危险状态，从而导致电气设备存在收到电力波动而波动的影响。所述专变采集终端用于对专变用户的多种电力数据中，用于进行电能质量评估的多种数据进行实时采集。所述数据采集源为对电力系统的动态电力数据进行监测的数据采集装置。
在本实施例中，根据所述监测指标集中评估电能质量所采用的数据项，确定对专变用户进行电能质量评估的多种类型监测数据，根据多种类型监测数据确定所述专变采集终端对专变用户进行数据采集的多个数据采集源。
S300：以所述数据采集源作为连接对象，确定多个数据传输通道；
S400：通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；
具体而言，在本实施例中，所述数据传输通道用于将数据采集设备在数据采集源获得的数据进行信息传递，送达电能质量监测系统进行电能质量评估。
所述数据传输通道与所述数据采集源具有对应关系，每一数据传输通道单一传输同一数据采集源的数据。
获取多个数据采集源在历史数据采集中存在的缺陷数据，基于数据采集源的历史缺陷数据获得与多个数据采集源对应的多个数据缺陷特征，基于多个数据缺陷特征遍历对应的多个数据采集源的实时输出数据进行数据特征识别，获得在实时采集数据中存在异常的所述异常特征集，所述异常特征集反映了当前基于数据采集源获得的实时采集数据当前存在的数据缺陷，所述异常特征集由多个数据采集源的异常数据构成。
S500：以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道；
进一步的，所述获取多个异常识别模型，本申请提供的方法步骤S500还包括：
S510：搭建传输质量评估模型，其中，所述传输质量评估模型与所述多个异常识别模型连接；
S520：将所述传输质量评估模型链接至所述多个数据传输通道，对所述多个数据传输通道进行传输质量评估，得到多个传输质量系数；
S530：将所述多个传输质量系数作为激活所述多个异常识别模型的约束条件。
具体而言，在本实施例中，对多个数据采集源对应设置多个异常识别模型，本实施例对于数据异常识别模型的构建和训练方法不作任何限制，将多个数据采集源在历史数据采集中存在的缺陷数据以及多个数据采集源的所述异常特征数据集作为训练数据进行模型训练，获得模型输出结果准确度达到了预设输出准确度要求的多个异常识别模型。根据多个异常识别模型与多个数据采集源的对应关系，将多个异常识别模型嵌入与多个数据采集源具有对应关系的所述多个数据传输通道。
同时，为避免数据采集设备所获数据在通道中必须经由异常识别模型识别处理，嵌入数据传输通道的异常识别模型在进行数据采集源数据传输过程中实时处于运行状态，导致对系统算力资源浪费以及降低数据传输效率。
在本实施例中，对所述数据异常识别模型设定激活条件，通过对实时采集数据在数据传输延迟以及数据损耗两个维度进行数据传输质量评估获得的数据传输质量系数，根据所获数据传输质量系数判断是否激活多个数据传输通道的多个异常识别模型。
具体的，在本实施例中，通过搭建所述传输质量评估模型在数据传输延迟和数据传输损耗两方面对实时采集数据进行数据传输质量评估。
将所述传输质量评估模型与所述多个异常识别模型连接，将所述传输质量评估模型链接至所述多个数据传输通道，对所述多个数据传输通道中传输的实时采集数据进行传输质量评估，得到多个传输质量系数，将所述多个传输质量系数作为激活所述多个异常识别模型的约束条件，将所述传输质量系数与多个异常识别模型预设的激活系数进行比对，确定对应需要激活，进行实时采集数据异常识别的异常识别模型。
本实施例通过在多个数据传输通道中对应嵌入多个异常识别模型，通过在数据传输通道中链入对数据传输质量进行评估的传输质量评估模型进行多个异常识别模型激活判断，达到了对于实时采集数据中异常数据进行准确识别同时避免异常数据识别对于数据传输效率的影响的技术效果。
S600：根据所述多个异常识别模型，输出异常识别结果；
具体而言，在本实施例中，所述异常识别结果为由数据传输通道进行数据传输过程中的产生的数据偏差，根据所述多个数据传输通道中嵌入的多个异常识别模型对应进行各个数据传输通道中数据的异常识别，获得所述异常识别结果，对所述异常识别结果进行数据还原和纠正，使各个数据传输通道中计划传输至电能质量检测系统中进行电能质量评估的各项相关数据具有可信性。
S700：将所述异常识别结果输入所述数据转换模块中，根据所述数据转换模块，得到数据转换结果；
进一步的，如图3所示，将所述异常识别结果输入所述数据转换模块中，根据所述数据转换模块，得到数据转换结果，本申请提供的方法步骤S700还包括：
S710：通过对所述异常识别结果进行异常源分析，获取标识异常源，其中，所述标识异常源为异常数据占比最大的数据源；
S720：以所述标识异常源进行模型训练，生成数据转换模型，其中，所述数据转换模型用于实现异常数据的标定纠正；
S730：以嵌入所述数据转换模型的所述数据转换模块，对所述异常识别结果进行纠正，输出所述数据转换结果。
具体而言，在本实施例中，所述数据转换模块用于将所述异常识别结果还原纠正为正常数据，所述标识异常源为异常数据中占比最大的数据源。
通过对所述异常识别结果进行异常源分析，确定引起数据异常的原因，例如某些设备采用远程数据传输或信号传输，存在信号干扰数据传输等数据异常原因。获取异常数据中占比最大的数据源作为所述标识异常源，所述标识异常源具有数据异常情况的代表性，以所述标识异常源进行模型训练，生成用于进行异常数据的标定纠正的数据转换模型。
本实施例对于所述数据转换模型的模型构建方法和训练方法不做限制，可根据实际需求进行设定。将训练好的所述数据转换模型嵌入所述数据转换模型的所述数据转换模块，对所述异常识别结果进行纠正，输出所述数据转换结果。
本实施例通过对多种类型的异常数据进行异常源分析，从中获取发生数据异常具有普遍性的数据源进行异常数据纠正模型的构建和训练，达到了对异常数据进行定向针对性纠正，提高数据传输通道传输质量，间接性提高电能质量评估结果的有效性和可信度的技术效果。
S800：将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。
具体而言，在本实施例中将所述数据转换结果和不存在数据异常的实时采集数据由所述多个数据传输通道传输至所述电能质量监测系统，由电能质量监测系统进行电能质量评估，输出电能质量评估结果。
本实施例提供的方法通过连接所述电能质量监测系统，获取用于进行电能质量评估的监测指标集；根据所述监测指标集，确定用于专变采集终端的数据采集源；为后续进行数据传输通道的构建提供参考基础，以所述数据采集源作为连接对象，确定多个数据传输通道；通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道，便于以数据通道为单位，对多类型数据进行异常识别，提高异常数据识别准确度和识别效率，根据所述多个异常识别模型，输出异常识别结果；将所述异常识别结果输入所述数据转换模块中，根据所述数据转换模块，得到数据转换结果，实现将异常数据纠正为正常数据，提高最终在电能质量监测模型中进行电能质量评估的参考数据的有效性；将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。达到了提高进行专变用户电能质量评估相关数据采集精度，提高专变用户电能质量评估结果在电力运维中的可参考性和有效性，实现确保专变用户的用电安全的技术效果。
进一步的，如图2所示，本申请提供的方法步骤还包括：
S210：获取所述数据采集源的数据采集设备的信息；
S220：通过对所述数据采集设备的信息进行设备工况数据提取，获取设备实时工况信息；
S230：根据所述设备实时工况信息，判断所述数据采集设备是否处于异常状态；
S240：若所述数据采集设备处于异常状态，对异常采集设备进行标识提醒。
进一步的，本申请提供的方法步骤还包括：
S231：判断所述数据采集设备是否为成套采集设备，其中，所述成套采集设备为辅助执行数据采集的子设备和母设备；
S232：所述数据采集设备为成套采集设备，获取所述成套采集设备的数据传输路径；
S233：以所述数据传输路径，确定数据传输异常源；
S234：根据所述数据传输异常源，定位所述异常采集设备。
具体而言，为确保基于所述数据采集源获得的用于进行专变用户电能质量评估的监测数据的有效性，避免由于进行数据采集的传感设备即数据采集设备异常导致的监测数据异常。本实施例在通过数据采集设备在数据采集源进行专变用户数据采集之前，验证当前数据采集设备是否存在状态异常。
在本实施例中，获取所述数据采集源的数据采集设备的信息，根据所述数据采集设备的信息获取所述数据采集设备的设备工况数据，将设备工况数据与设备工况数据采集实际作为所述获取设备实时工况信息，根据所述设备实时工况信息，生成工况数据波动曲线图像，根据数据波动曲线变化是否规律判断所述数据采集设备是否处于异常状态。
若所述数据采集设备处于异常状态，则进一步判断所述数据采集设备的设备组成类型，数据采集设备的设备组成类型包括单一数据采集设备和成套采集设备，单一数据采集设备为基于单个设备进行数据源数据采集的设备，成套数据采集设备为通过包括辅助执行数据采集的子设备和母设备配合进行数据采集的组合设备，所述成套数据采集设备中的数据传输路径为子设备传递至母设备，母设备与子设备之间为一一对应关系或一多对应关系。
当所述数据采集设备为单一采集设备时，则对所述数据采集设备进行设备异常标识提醒，停止该数据采集设备的数据采集运行。
当所述数据采集设备为成套采集设备时，则进一步确定所述成套采集设备中辅助执行数据采集的子设备和母设备，获取所述成套采集设备中具体配合进行数据传输的子设备和母设备，获得所述数据传输路径，根据所述数据传输路径，确定数据传输异常源在数据传输过程中还是在数据传输来源处，根据所述数据传输异常源，定位所述异常采集设备，当数据传输异常源在数据传输过程中，则定位异常采集母设备，当数据传输异常源在数据传输来源处，则对应定位异常采集子设备，对异常采集设备进行标识提醒，用于后续进行数据源数据采集时的数据筛除以避免错误数据流入电能质量监测系统导致电能质量评估结果可参考性降低。
本实施例通过在执行数据源数据采集之前，对用于进行数据采集的设备进行运行工况分析，对存在异常工况的数据采集设备进行标识，避免异常数据流入电能质量检测系统参与电能质量评估，达到了提高专变用户电能质量评估结果有效性和可参考性的技术效果。
进一步的，以所述数据采集源作为连接对象，确定多个数据传输通道之后，本申请提供的方法步骤S520还包括：
S521：根据所述多个数据传输通道，确定多组测试样本，其中，所述多组测试样本与所述多个数据传输通道对应，且每组测试样本中包括多组数据集；
S522：基于所述多个测试样本对所述多个数据传输通道进行传输帧周期测试，输出多个传输延迟指数，其中，所述多个传输延迟指数与所述多个数据传输通道一一对应；
S523：将所述多个传输延迟指数作为第一数据集进行计算，生成所述多个传输质量系数。
进一步的，生成所述多个传输质量系数，本申请提供的方法步骤S523还包括：
S523-1：基于所述多个测试样本，获取经所述多个数据传输通道得到的多个输出样本；
S523-2：以所述多个测试样本和所述多个输出样本进行数据损耗分析，输出多个数据损耗指数，其中，所述多个数据损耗指数与所述多个数据传输通道一一对应；
S523-3：将所述多个数据损耗指数作为第二数据集进行计算，生成所述多个传输质量系数。
具体而言，所述传输帧周期测试为根据传输数据从数据采集设备作为发送端发送的时间节点到传输数据经由数据传输通道抵达电能质量监测系统作为接收端接受传输数据的响应时长进行传输周期测试生成的传输帧周期。
在本实施例中，根据所述多个数据传输通道，确定多组测试样本，所述多组测试样本的数据类型与所述多个数据传输通道的传输数据类型对应，且每组测试样本中包括多组数据大小不同的数据集。
基于所述多个测试样本对所述多个数据传输通道进行传输帧周期测试，根据多个测试样本在多个数据传输通道中的数据传输帧周期输出多个传输延迟指数，所述传输延迟指数以时间为单位。
基于所述多个测试样本，获取经所述多个数据传输通道得到的多个输出样本，根据所述多个测试样本和所述多个输出样本进行数据损耗分析，根据测试样本与输出样本之间的数据损失程度输出多个数据损耗指数，所述数据损耗指数以数据占用空间为单位，所述多个数据损耗指数与所述多个数据传输通道一一对应。
将所述多个传输延迟指数和所述多个数据损耗指数进行归一化处理，以所述多个传输延迟指数作为第一数据集，以所述多个数据损耗指数作为第二数据集进行计算，生成所述多个传输质量系数，所述多个传输质量系数与所述多个数据传输通道一一对应。
本实施例通过构建传输质量评估模型链入多个数据传输通道，用于在数据传输延迟和数据传输损耗两个维度评估判断是否需要激活嵌入各个数据传输通道中的异常识别模型，达到了降低异常识别模型对于系统算力资源的耗费，以及避免所有进行数据传输通道的数据都需要执行异常数据识别分析引起的数据传输效率降低，导致电能质量监测模型获得的电能质量时效性降低的技术效果。
实施例二
基于与前述实施例中一种基于专变采集终端的电能质量监测方法相同的发明构思，如图4所示，本申请提供了一种基于专变采集终端的电能质量监测系统，其中，所述系统包括：
检测指标获得模块11，用于连接电能质量监测系统，获取用于进行电能质量评估的监测指标集；
数据来源确定模块12，用于根据所述监测指标集，确定用于专变采集终端的数据采集源；
数据通道确定模块13，用于以所述数据采集源作为连接对象，确定多个数据传输通道；
数据特征识别模块14，用于通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；
识别模型训练模块15，用于以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道；
异常识别输出模块16，用于根据所述多个异常识别模型，输出异常识别结果；
数据转换执行模块17，用于将所述异常识别结果输入数据转换模块中，根据所述数据转换模块，得到数据转换结果；
电能质量评估模块18，用于将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。
进一步的，所述数据来源确定模块12还包括：
设备信息获得单元，用于获取所述数据采集源的数据采集设备的信息；
设备工况采集单元，用于通过对所述数据采集设备的信息进行设备工况数据提取，获取设备实时工况信息；
设备状态判断单元，用于根据所述设备实时工况信息，判断所述数据采集设备是否处于异常状态；
设备异常标识单元，用于若所述数据采集设备处于异常状态，对异常采集设备进行标识提醒。
进一步的，所述设备状态判断单元还包括：
设备构成判断单元，用于判断所述数据采集设备是否为成套采集设备，其中，所述成套采集设备为辅助执行数据采集的子设备和母设备；
数据传输判断单元，用于所述数据采集设备为成套采集设备，获取所述成套采集设备的数据传输路径；
异常源头确定单元，用于以所述数据传输路径，确定数据传输异常源；
异常设备定位单元，用于根据所述数据传输异常源，定位所述异常采集设备。
进一步的，所述识别模型训练模块15还包括：
评估模型搭建单元，用于搭建传输质量评估模型，其中，所述传输质量评估模型与所述多个异常识别模型连接；
评估模型链接单元，用于将所述传输质量评估模型链接至所述多个数据传输通道，对所述多个数据传输通道进行传输质量评估，得到多个传输质量系数；
约束条件预设单元，用于将所述多个传输质量系数作为激活所述多个异常识别模型的约束条件。
进一步的，所述数据通道确定模块13还包括：
测试样本确定单元，用于根据所述多个数据传输通道，确定多组测试样本，其中，所述多组测试样本与所述多个数据传输通道对应，且每组测试样本中包括多组数据集；
传输周期测试单元，用于基于所述多个测试样本对所述多个数据传输通道进行传输帧周期测试，输出多个传输延迟指数，其中，所述多个传输延迟指数与所述多个数据传输通道一一对应；
传输质量计算单元，用于将所述多个传输延迟指数作为第一数据集进行计算，生成所述多个传输质量系数。
进一步的，所述传输质量计算单元还包括：
输出样本获得单元，用于基于所述多个测试样本，获取经所述多个数据传输通道得到的多个输出样本；
数据损耗分析单元，用于以所述多个测试样本和所述多个输出样本进行数据损耗分析，输出多个数据损耗指数，其中，所述多个数据损耗指数与所述多个数据传输通道一一对应；
传输质量生成单元，用于将所述多个数据损耗指数作为第二数据集进行计算，生成所述多个传输质量系数。
进一步的，所述数据转换执行模块17还包括：
异常源分析单元，用于通过对所述异常识别结果进行异常源分析，获取标识异常源，其中，所述标识异常源为异常数据占比最大的数据源；
异常数据纠正单元，用于以所述标识异常源进行模型训练，生成数据转换模型，其中，所述数据转换模型用于实现异常数据的标定纠正；
数据转换输出单元，用于以嵌入所述数据转换模型的所述数据转换模块，对所述异常识别结果进行纠正，输出所述数据转换结果。
综上所述的任意一项方法或者步骤可作为计算机指令或程序存储在各种不限类型的计算机存储器中，通过各种不限类型的计算机处理器识别计算机指令或程序，进而实现上述任一项方法或者步骤。
基于本发明的上述具体实施例，本技术领域的技术人员在不脱离本发明原理的前提下，对本发明所作的任何改进和修饰，皆应落入本发明的专利保护范围。

Claims (8)

  1. 一种基于专变采集终端的电能质量监测方法，其特征在于，所述方法应用于电能质量监测系统，所述系统与数据转换模块通信连接，所述方法包括：
    连接所述电能质量监测系统，获取用于进行电能质量评估的监测指标集；
    根据所述监测指标集，确定用于专变采集终端的数据采集源；
    以所述数据采集源作为连接对象，确定多个数据传输通道；
    通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；
    以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道；
    根据所述多个异常识别模型，输出异常识别结果；
    将所述异常识别结果输入所述数据转换模块中，根据所述数据转换模块，得到数据转换结果；
    将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。
  2. 如权利要求1所述的方法，其特征在于，所述方法还包括：
    获取所述数据采集源的数据采集设备的信息；
    通过对所述数据采集设备的信息进行设备工况数据提取，获取设备实时工况信息；
    根据所述设备实时工况信息，判断所述数据采集设备是否处于异常状态；
    若所述数据采集设备处于异常状态，对异常采集设备进行标识提醒。
  3. 如权利要求2所述的方法，其特征在于，所述方法还包括：
    判断所述数据采集设备是否为成套采集设备，其中，所述成套采集设备为辅助执行数据采集的子设备和母设备；
    所述数据采集设备为成套采集设备，获取所述成套采集设备的数据传输路径；
    以所述数据传输路径，确定数据传输异常源；
    根据所述数据传输异常源，定位所述异常采集设备。
  4. 如权利要求1所述的方法，其特征在于，所述获取多个异常识别模型，所述方法还包括：
    搭建传输质量评估模型，其中，所述传输质量评估模型与所述多个异常识别模型连接；
    将所述传输质量评估模型链接至所述多个数据传输通道，对所述多个数据传输通道进行传输质量评估，得到多个传输质量系数；
    将所述多个传输质量系数作为激活所述多个异常识别模型的约束条件。
  5. 如权利要求4所述的方法，其特征在于，以所述数据采集源作为连接对象，确定多个数据传输通道之后，所述方法还包括：
    根据所述多个数据传输通道，确定多组测试样本，其中，所述多组测试样本与所述多个数据传输通道对应，且每组测试样本中包括多组数据集；
    基于所述多个测试样本对所述多个数据传输通道进行传输帧周期测试，输出多个传输延迟指数，其中，所述多个传输延迟指数与所述多个数据传输通道一一对应；
    将所述多个传输延迟指数作为第一数据集进行计算，生成所述多个传输质量系数。
  6. 如权利要求5所述的方法，其特征在于，生成所述多个传输质量系数，所述方法还包括：
    基于所述多个测试样本，获取经所述多个数据传输通道得到的多个输出样本；
    以所述多个测试样本和所述多个输出样本进行数据损耗分析，输出多个数据损耗指数，其中，所述多个数据损耗指数与所述多个数据传输通道一一对应；
    将所述多个数据损耗指数作为第二数据集进行计算，生成所述多个传输质量系数。
  7. 如权利要求1所述的方法，其特征在于，将所述异常识别结果输入所述数据转换模块中，根据所述数据转换模块，得到数据转换结果，所述方法还包括：
    通过对所述异常识别结果进行异常源分析，获取标识异常源，其中，所述标识异常源为异常数据占比最大的数据源；
    以所述标识异常源进行模型训练，生成数据转换模型，其中，所述数据转换模型用于实现异常数据的标定纠正；
    以嵌入所述数据转换模型的所述数据转换模块，对所述异常识别结果进行纠正，输出所述数据转换结果。
  8. 一种基于专变采集终端的电能质量监测系统，其特征在于，所述系统包括：
    检测指标获得模块，用于连接电能质量监测系统，获取用于进行电能质量评估的监测指标集；
    数据来源确定模块，用于根据所述监测指标集，确定用于专变采集终端的数据采集源；
    数据通道确定模块，用于以所述数据采集源作为连接对象，确定多个数据传输通道；
    数据特征识别模块，用于通过对所述数据采集源进行数据特征识别，获取异常特征集，其中，所述异常特征集为在实时采集数据中存在异常的数据集合；
    识别模型训练模块，用于以所述异常特征集作为训练集进行模型训练，获取多个异常识别模型，其中，所述多个异常识别模型分别嵌于所述多个数据传输通道；
    异常识别输出模块，用于根据所述多个异常识别模型，输出异常识别结果；
    数据转换执行模块，用于将所述异常识别结果输入数据转换模块中，根据所述数据转换模块，得到数据转换结果；
    电能质量评估模块，用于将所述数据转换结果由所述多个数据传输通道传输至所述电能质量监测系统，用于进行电能质量评估。
PCT/CN2023/121728 2022-11-11 2023-09-26 一种基于专变采集终端的电能质量监测方法及系统 WO2024098990A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211411814.5A CN115453254B (zh) 2022-11-11 2022-11-11 一种基于专变采集终端的电能质量监测方法及系统
CN202211411814.5 2022-11-11

Publications (1)

Publication Number Publication Date
WO2024098990A1 true WO2024098990A1 (zh) 2024-05-16

Family

ID=84295578

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/121728 WO2024098990A1 (zh) 2022-11-11 2023-09-26 一种基于专变采集终端的电能质量监测方法及系统

Country Status (2)

Country Link
CN (1) CN115453254B (zh)
WO (1) WO2024098990A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115453254B (zh) * 2022-11-11 2023-05-02 浙江万胜智能科技股份有限公司 一种基于专变采集终端的电能质量监测方法及系统
CN117239747B (zh) * 2023-11-16 2024-02-06 江苏濠汉信息技术有限公司 一种基于模型识别的宿舍安全用电的控制方法及系统
CN117408537B (zh) * 2023-12-15 2024-05-07 安徽科派自动化技术有限公司 一种能够实现实时风险预测的电能质量监测系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102053202A (zh) * 2009-11-10 2011-05-11 北京博电新力电力系统仪器有限公司 面向智能电网的电能质量监测系统及方法
KR20110100855A (ko) * 2010-03-05 2011-09-15 가톨릭대학교 산학협력단 스마트 그리드 전력품질 원격 모니터링 방법 및 시스템
CN110824282A (zh) * 2019-11-20 2020-02-21 湖南铁路科技职业技术学院 一种高压充电桩电能质量监测系统
CN111198979A (zh) * 2019-12-31 2020-05-26 中国电力科学研究院有限公司 一种用于对输变电可靠性评估大数据进行清洗的方法及系统
CN112000672A (zh) * 2020-08-25 2020-11-27 杭州电力设备制造有限公司 一种分布式电源电能质量监测系统
CN112949721A (zh) * 2021-03-04 2021-06-11 吴统明 一种电力设备静态数据质量评估方法及系统
US20220036137A1 (en) * 2018-09-19 2022-02-03 Rulex, Inc. Method for detecting anomalies in a data set
CN115453254A (zh) * 2022-11-11 2022-12-09 浙江万胜智能科技股份有限公司 一种基于专变采集终端的电能质量监测方法及系统

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013040000A2 (en) * 2011-09-12 2013-03-21 United Parcel Service Of America, Inc. Service exception analysis systems and methods
WO2016138750A1 (zh) * 2015-03-04 2016-09-09 江苏省电力公司常州供电公司 一种电能质量扰动源定位系统及定位方法
KR101779707B1 (ko) * 2016-03-31 2017-09-19 전자부품연구원 전자소자가 실장된 전자모듈의 수명정량평가방법
US10467879B2 (en) * 2017-10-19 2019-11-05 Google Llc Thoughtful elderly monitoring in a smart home environment
JP7252542B2 (ja) * 2019-05-21 2023-04-05 本田技研工業株式会社 覚醒状態推定装置及び覚醒状態推定方法
CN111652570A (zh) * 2020-05-12 2020-09-11 国网福建省电力有限公司 一种自动审核系统及方法
CN112288594A (zh) * 2020-10-23 2021-01-29 国网辽宁省电力有限公司信息通信分公司 一种基于实时事件触发的数据质量异动处理方法和系统
CN112214481A (zh) * 2020-10-27 2021-01-12 深圳供电局有限公司 一种电能质量数据的数据清洗方法
CN112418053B (zh) * 2020-11-18 2023-12-26 华北电力大学 一种基于递进式识别来监测异常状态的系统及方法
CN112394248A (zh) * 2020-11-20 2021-02-23 南方电网数字电网研究院有限公司 电能质量测量装置、系统和方法
CN114089033B (zh) * 2022-01-24 2022-04-26 天津安力信通讯科技有限公司 一种基于频谱分析的异常信号检测方法及系统
CN114331761B (zh) * 2022-03-15 2022-07-08 浙江万胜智能科技股份有限公司 一种专变采集终端的设备参数分析调整方法及系统
CN114640177B (zh) * 2022-03-24 2022-10-04 重庆伏特猫科技有限公司 一种基于电力能效监测装置的电力能效监测方法
CN115037603A (zh) * 2022-05-31 2022-09-09 国网湖南省电力有限公司 用电信息采集设备的诊断评估方法、装置及系统
CN115080290B (zh) * 2022-06-07 2023-07-07 吉林大学 一种基于智能算法的异常数据检测方法及系统
CN115296905B (zh) * 2022-08-04 2024-06-04 新疆品宣生物科技有限责任公司 一种基于移动终端的数据采集分析方法及系统
CN115170000B (zh) * 2022-09-06 2023-01-13 浙江万胜智能科技股份有限公司 一种基于电能表通信模块的远程监测方法及系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102053202A (zh) * 2009-11-10 2011-05-11 北京博电新力电力系统仪器有限公司 面向智能电网的电能质量监测系统及方法
KR20110100855A (ko) * 2010-03-05 2011-09-15 가톨릭대학교 산학협력단 스마트 그리드 전력품질 원격 모니터링 방법 및 시스템
US20220036137A1 (en) * 2018-09-19 2022-02-03 Rulex, Inc. Method for detecting anomalies in a data set
CN110824282A (zh) * 2019-11-20 2020-02-21 湖南铁路科技职业技术学院 一种高压充电桩电能质量监测系统
CN111198979A (zh) * 2019-12-31 2020-05-26 中国电力科学研究院有限公司 一种用于对输变电可靠性评估大数据进行清洗的方法及系统
CN112000672A (zh) * 2020-08-25 2020-11-27 杭州电力设备制造有限公司 一种分布式电源电能质量监测系统
CN112949721A (zh) * 2021-03-04 2021-06-11 吴统明 一种电力设备静态数据质量评估方法及系统
CN115453254A (zh) * 2022-11-11 2022-12-09 浙江万胜智能科技股份有限公司 一种基于专变采集终端的电能质量监测方法及系统

Also Published As

Publication number Publication date
CN115453254A (zh) 2022-12-09
CN115453254B (zh) 2023-05-02

Similar Documents

Publication Publication Date Title
WO2024098990A1 (zh) 一种基于专变采集终端的电能质量监测方法及系统
CN116660669B (zh) 一种电力设备故障在线监测系统及方法
CN109029582A (zh) 一种基于多种参数采集的电力变压器故障在线检测系统
CN110311709B (zh) 用电信息采集系统故障判别方法
CN104090252B (zh) 一种电缆型故障指示器的综合检测系统和方法
CN113532598B (zh) 智能水表在线故障检测调节系统
CN114255784A (zh) 一种基于声纹识别的变电站设备故障诊断方法及相关装置
CN116779119A (zh) 一种适用于血液内科的血液存储环境监管系统
CN113177646A (zh) 一种基于自适应边缘代理的配电设备在线监测方法及系统
CN201583643U (zh) 一种电能质量分析仪全自动检定、校准装置
CN108377209A (zh) 基于scada的设备故障检测系统和检测方法
WO2024098986A1 (zh) 一种基于智能录波器的继电保护装置缺陷检测方法及系统
CN116256691B (zh) 一种电能表失准在线监测方法及系统
TW201017447A (en) An early stage abnormality prediction device for fabrication process and its prediction method
CN111626542A (zh) 一种用于发电机励磁系统的基于大数据的检测方法
CN116566839A (zh) 一种电力企业通信资源质量评估系统
CN116191677A (zh) 一种变电站智能监控管理方法及系统
CN116611953A (zh) 基于物联网的电能表生产制造实时数据管理方法和系统
CN113591909B (zh) 电力系统的异常检测方法、异常检测装置以及存储介质
CN114049697B (zh) 一种巡点检辅助系统
CN105550083A (zh) Mss自动化测试方法及系统
CN106059093A (zh) 智能变电站健康状态监测系统
CN115060998A (zh) 一种输电线路用电异常的确定方法、装置、设备及介质
CN106646019A (zh) 采用二次状态检测的继电自动化测试仪及闭环测试方法
EP3388912A1 (en) Method and system for determining and reporting equipment operating conditions and health status

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: 23887683

Country of ref document: EP

Kind code of ref document: A1