技术领域Technical Field
本发明涉及数据处理技术领域，并且更具体地，涉及一种继电保护缺陷数据的唯一性校核方法及系统。The present invention relates to the technical field of data processing, and more specifically, to a uniqueness verification method and system for relay protection defect data.
背景技术Background technique
继电保护数据可划分为设备管理、运行维护、状态评价和定值整定等类型。设备管理类数据包含保护设备台账以及台账与一次设备、厂站间关系。定值整定类数据包含保护定值模型、定值项、保护定值同系统运行方式的关系等。状态评价类数据模型包含基于智能变电站配置文件(SCD文件)所生成的IED设备信息，以及每一台IED设备所对应的状态量、告警、动作以及模拟量等信息模型。运行维护类数据包含巡检巡视、缺陷、家族性缺陷数据等。继电保护缺陷数据是一类描述继电保护设备运行异常及故障的设备运行维护数据，能够体现出设备运行可靠性、不同类型设备总体质量等信息。Relay protection data can be divided into types such as equipment management, operation and maintenance, status evaluation and set value setting. Equipment management data includes protection equipment ledgers and the relationship between the ledgers and primary equipment and plants. Set value setting data includes protection set value models, set value items, and the relationship between protection set values and system operation modes. The status evaluation data model includes IED device information generated based on the smart substation configuration file (SCD file), as well as information models such as state quantities, alarms, actions, and analog quantities corresponding to each IED device. Operation and maintenance data includes inspection and patrol, defects, and family defect data. Relay protection defect data is a type of equipment operation and maintenance data that describes abnormal operation and failures of relay protection equipment, which can reflect information such as equipment operation reliability and the overall quality of different types of equipment.
目前，对继电保护专业信息的管理和应用还停留在单一信息的采集应用层面，数据的准确性、完整性、标准性方面问题突出，尚未实现数据从保护装置源端至变电站端、调度端的信息自动采集及智能录入管理。因此，亟需开展继电保护装置标识技术和数据校核技术的研究工作，实现全网保护设备数据的唯一识别，全面完善保护设备全寿命周期信息模型，为继电保护信息准确、完备采集及分析应用奠定基础，实现继电保护专业各应用系统基于标识信息的关联应用。At present, the management and application of relay protection professional information is still at the level of single information collection and application, with prominent problems in data accuracy, completeness and standardization, and the automatic collection and intelligent entry management of data from the source of protection devices to the substation and dispatching ends has not yet been realized. Therefore, it is urgent to carry out research on relay protection device identification technology and data verification technology, realize the unique identification of protection equipment data across the entire network, comprehensively improve the information model of the entire life cycle of protection equipment, lay the foundation for accurate and complete collection and analysis of relay protection information, and realize the associated application of various application systems of relay protection based on identification information.
其中，继电保护数据的唯一性是数据治理的重要方面。继电保护专业系统中常存在重复的继电保护缺陷记录，影响了缺陷数据的应用水平。Among them, the uniqueness of relay protection data is an important aspect of data governance. There are often duplicate relay protection defect records in the relay protection professional system, which affects the application level of defect data.
发明内容Summary of the invention
本发明提出一种继电保护缺陷数据的唯一性校核方法及系统，以解决如何自动对继电保护缺陷数据进行唯一性校核的问题。The present invention provides a uniqueness checking method and system for relay protection defect data, so as to solve the problem of how to automatically check the uniqueness of relay protection defect data.
为了解决上述问题，根据本发明的一个方面，提供了一种继电保护缺陷数据的唯一性校核方法，所述方法包括：In order to solve the above problem, according to one aspect of the present invention, a method for checking the uniqueness of relay protection defect data is provided, the method comprising:
获取同一继电保护设备标识对应的至少两条待校验的缺陷记录，并选取所述至少两条待校验的缺陷记录中的两条待校验的缺陷记录作为待校验的缺陷记录对；Acquire at least two defect records to be verified corresponding to the same relay protection device identifier, and select two defect records to be verified from the at least two defect records to be verified as a defect record pair to be verified;
根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，并根据所述待校验的缺陷记录对中的缺陷属性数据以及每项缺陷属性的权重计算所述待校验的缺陷记录对的缺陷属性差异度The weight of each defect attribute is calculated according to the statistical results of each attribute of all defect records of the relay protection, and the defect attribute difference degree of the defect record pair to be verified is calculated according to the defect attribute data in the defect record pair to be verified and the weight of each defect attribute.
根据所述待校验的缺陷记录对中的缺陷时间确定所述待校验的缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度；Determining the defect time difference and defect time span separation of the defect record pair to be verified according to the defect time in the defect record pair to be verified;
获取多个样本缺陷记录对，计算每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度，并根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面；Acquire multiple sample defect record pairs, calculate the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair, and use support vector machine to determine the defect record uniqueness verification interface based on the data points composed of the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair;
根据所述待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度对应的数据点与所述缺陷记录唯一性校核分界面的相对位置，确定所述待校验的缺陷记录对的唯一性校核结果；其中，所述唯一性校核结果包括：重复缺陷记录对和非重复缺陷记录对。According to the relative positions of the data points corresponding to the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be verified and the defect record uniqueness verification interface, the uniqueness verification result of the defect record pair to be verified is determined; wherein the uniqueness verification result includes: duplicate defect record pairs and non-duplicate defect record pairs.
优选地，其中所述方法利用如下方式根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，包括：Preferably, the method calculates the weight of each defect attribute according to the statistical results of various attributes of all defect records of relay protection in the following manner, including:
其中，ωi为第i项缺陷属性的权重；Pij为第i项缺陷属性的第j个取值出现的概率；k为第i项缺陷属性的不同取值的个数。Among them,ωi is the weight of the i-th defect attribute;Pij is the probability of the j-th value of the i-th defect attribute occurring; and k is the number of different values of the i-th defect attribute.
优选地，其中所述方法利用如下公式根据每项缺陷属性的权重计算所述缺陷记录对的缺陷属性差异度，包括：Preferably, the method calculates the defect attribute difference of the defect record pair according to the weight of each defect attribute using the following formula, including:
其中，V1为缺陷记录对的缺陷属性差异度；ωi为第i项缺陷属性的权重；Ii表示缺陷记录对第i项缺陷属性的数据是否有差别，如果有差别则取1，无差别则取0。Wherein,V1 is the difference degree of defect attribute of the defect record pair;ωi is the weight of the i-th defect attribute;Ii indicates whether the data of the defect record pair i-th defect attribute are different, if there is a difference, it takes 1, if there is no difference, it takes 0.
优选地，其中所述方法利用如下方式根据缺陷记录对中的缺陷时间确定所述缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度，包括：Preferably, the method determines the defect time difference and defect time span separation of the defect record pair according to the defect time in the defect record pair in the following manner, including:
V2＝(V21+V22)/2，V2 = (V21 +V22 ) / 2,
其中，V2为缺陷记录对的缺陷时间差异度；V21为缺陷记录对的缺陷发生时间差异度；V22为缺陷记录对的缺陷消除时间差异度；Tdif1和Tdif2分别为缺陷记录对中的缺陷发生时间之差和缺陷消除时间之差，单位为分钟；V3为缺陷记录对的缺陷时间跨度分隔度。Among them,V2 is the defect time difference of the defect record pair;V21 is the defect occurrence time difference of the defect record pair;V22 is the defect elimination time difference of the defect record pair;Tdif1 andTdif2 are the difference in defect occurrence time and defect elimination time in the defect record pair, respectively, in minutes;V3 is the defect time span separation of the defect record pair.
优选地，其中所述根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面，包括：Preferably, the method of determining the defect record uniqueness verification interface using a support vector machine based on the data points consisting of the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair includes:
构建以缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度为坐标的指标空间；Construct an index space with defect attribute difference, defect time difference and defect time span separation as coordinates;
在所述指标空间中确定每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，并根据每个样本缺陷记录对的是否为重复缺陷记录的先验结果进行分簇，确定重复缺陷记录簇和非重复缺陷记录簇；Determine the data points consisting of defect attribute difference, defect time difference and defect time span separation of each sample defect record pair in the index space, and perform clustering according to the prior result of whether each sample defect record pair is a duplicate defect record to determine a duplicate defect record cluster and a non-duplicate defect record cluster;
根据所述重复缺陷簇和非重复缺陷簇，采用支持向量机进行训练，以确定缺陷记录唯一性校核分界面。According to the repeated defect clusters and the non-repeated defect clusters, a support vector machine is used for training to determine a defect record uniqueness verification interface.
根据本发明的另一个方面，提供了一种继电保护缺陷数据的唯一性校核系统，所述系统包括：According to another aspect of the present invention, a uniqueness verification system for relay protection defect data is provided, the system comprising:
待校验的缺陷记录对确定单元，用于获取同一继电保护设备标识对应的至少两条待校验的缺陷记录，并选取所述至少两条待校验的缺陷记录中的两条待校验的缺陷记录作为待校验的缺陷记录对；A defect record pair determination unit to be verified, used for obtaining at least two defect records to be verified corresponding to the same relay protection device identifier, and selecting two defect records to be verified from the at least two defect records to be verified as the defect record pair to be verified;
缺陷属性差异度确定单元，用于根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，并根据所述待校验的缺陷记录对中的缺陷属性数据以及每项缺陷属性的权重计算所述待校验的缺陷记录对的缺陷属性差异度；A defect attribute difference determination unit, used to calculate the weight of each defect attribute according to the statistical results of various attributes of all defect records of relay protection, and calculate the defect attribute difference of the defect record pair to be verified according to the defect attribute data in the defect record pair to be verified and the weight of each defect attribute;
缺陷时间差异度确定单元，用于根据所述待校验的缺陷记录对中的缺陷时间确定所述待校验的缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度；A defect time difference determination unit, used to determine the defect time difference and defect time span separation of the defect record pair to be verified according to the defect time in the defect record pair to be verified;
缺陷记录唯一性校核分界面确定单元，用于获取多个样本缺陷记录对，计算每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度，并根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面；A defect record uniqueness verification interface determination unit is used to obtain multiple sample defect record pairs, calculate the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair, and determine the defect record uniqueness verification interface using a support vector machine based on the data points consisting of the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair;
唯一性校核结果确定单元，用于根据所述待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度对应的数据点与所述缺陷记录唯一性校核分界面的相对位置，确定所述待校验的缺陷记录对的唯一性校核结果；其中，所述唯一性校核结果包括：重复缺陷记录对和非重复缺陷记录对。A uniqueness verification result determination unit is used to determine the uniqueness verification result of the defect record pair to be verified based on the relative position of the data points corresponding to the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be verified and the defect record uniqueness verification interface; wherein the uniqueness verification result includes: duplicate defect record pairs and non-duplicate defect record pairs.
优选地，其中所述系统利用如下方式根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，包括：Preferably, the system calculates the weight of each defect attribute according to the statistical results of various attributes of all defect records of relay protection in the following manner, including:
其中，ωi为第i项缺陷属性的权重；Pij为第i项缺陷属性的第j个取值出现的概率；k为第i项缺陷属性的不同取值的个数。Among them,ωi is the weight of the i-th defect attribute;Pij is the probability of the j-th value of the i-th defect attribute occurring; and k is the number of different values of the i-th defect attribute.
优选地，其中所述系统利用如下公式根据每项缺陷属性的权重计算所述缺陷记录对的缺陷属性差异度，包括：Preferably, the system calculates the defect attribute difference of the defect record pair according to the weight of each defect attribute using the following formula, including:
其中，V1为缺陷记录对的缺陷属性差异度；ωi为第i项缺陷属性的权重；Ii表示缺陷记录对第i项缺陷属性的数据是否有差别，如果有差别则取1，无差别则取0。Wherein,V1 is the difference degree of defect attribute of the defect record pair;ωi is the weight of the i-th defect attribute;Ii indicates whether the data of the defect record pair i-th defect attribute are different, if there is a difference, it takes 1, if there is no difference, it takes 0.
优选地，其中所述系统利用如下方式根据缺陷记录对中的缺陷时间确定所述缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度，包括：Preferably, the system determines the defect time difference and defect time span separation of the defect record pair according to the defect time in the defect record pair in the following manner, including:
V2＝(V21+V22)/2，V2 = (V21 +V22 ) / 2,
其中，V2为缺陷记录对的缺陷时间差异度；V21为缺陷记录对的缺陷发生时间差异度；V22为缺陷记录对的缺陷消除时间差异度；Tdif1和Tdif2分别为缺陷记录对中的缺陷发生时间之差和缺陷消除时间之差，单位为分钟；V3为缺陷记录对的缺陷时间跨度分隔度。Among them,V2 is the defect time difference of the defect record pair;V21 is the defect occurrence time difference of the defect record pair;V22 is the defect elimination time difference of the defect record pair;Tdif1 andTdif2 are the difference in defect occurrence time and defect elimination time in the defect record pair, respectively, in minutes;V3 is the defect time span separation of the defect record pair.
优选地，其中所述缺陷记录唯一性校核分界面确定单元，根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面，包括：Preferably, the defect record uniqueness verification interface determination unit determines the defect record uniqueness verification interface using a support vector machine according to data points consisting of defect attribute difference, defect time difference and defect time span separation of each sample defect record pair, including:
构建以缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度为坐标的指标空间；Construct an index space with defect attribute difference, defect time difference and defect time span separation as coordinates;
在所述指标空间中确定每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，并根据每个样本缺陷记录对的是否为重复缺陷记录的先验结果进行分簇，确定重复缺陷记录簇和非重复缺陷记录簇；Determine the data points consisting of defect attribute difference, defect time difference and defect time span separation of each sample defect record pair in the index space, and perform clustering according to the prior result of whether each sample defect record pair is a duplicate defect record to determine a duplicate defect record cluster and a non-duplicate defect record cluster;
根据所述重复缺陷簇和非重复缺陷簇，采用支持向量机进行训练，以确定缺陷记录唯一性校核分界面。According to the repeated defect clusters and the non-repeated defect clusters, a support vector machine is used for training to determine a defect record uniqueness verification interface.
本发明提供了一种继电保护缺陷数据的唯一性校核方法及系统，计算待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度；根据多个样本缺陷记录对，利用支持向量机确定缺陷记录唯一性校核分界面；根据所述待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度对应的数据点与所述缺陷记录唯一性校核分界面的相对位置，确定所述待校验的缺陷记录对的唯一性校核结果；本发明能够有效支撑继电保护专业管理人员分析继电保护缺陷，能够准确地判断同一设备的任意2次缺陷记录是否重复，能够减轻人工分析数据的工作量，减小了人力物力成本，通过筛查、清理重复的缺陷数据，能够提升继电保护缺陷数据的综合应用水平。The present invention provides a uniqueness verification method and system for relay protection defect data, which calculate the defect attribute difference, defect time difference and defect time span separation of a defect record pair to be verified; determine the defect record uniqueness verification interface using a support vector machine based on multiple sample defect record pairs; determine the uniqueness verification result of the defect record pair to be verified based on the relative position of the data points corresponding to the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be verified and the defect record uniqueness verification interface; the present invention can effectively support professional managers of relay protection to analyze relay protection defects, can accurately determine whether any two defect records of the same equipment are repeated, can reduce the workload of manual data analysis, and reduce the cost of manpower and material resources. By screening and cleaning up repeated defect data, the comprehensive application level of relay protection defect data can be improved.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
通过参考下面的附图，可以更为完整地理解本发明的示例性实施方式：A more complete understanding of exemplary embodiments of the present invention may be obtained by referring to the following drawings:
图1为根据本发明实施方式的继电保护缺陷数据的唯一性校核方法100的流程图；FIG1 is a flow chart of a uniqueness checking method 100 for relay protection defect data according to an embodiment of the present invention;
图2为根据本发明实施方式的确定两条缺陷记录的缺陷时间跨度的交集和并集的示意图；FIG2 is a schematic diagram of determining the intersection and union of defect time spans of two defect records according to an embodiment of the present invention;
图3为根据本发明实施方式的支持向量机的示意图；FIG3 is a schematic diagram of a support vector machine according to an embodiment of the present invention;
图4为根据本发明实施方式的缺陷记录唯一性校核分界面的示意图；FIG4 is a schematic diagram of a defect record uniqueness verification interface according to an embodiment of the present invention;
图5为根据本发明实施方式的继电保护缺陷数据的唯一性校核系统500的结构示意图。FIG5 is a schematic structural diagram of a uniqueness checking system 500 for relay protection defect data according to an embodiment of the present invention.
具体实施方式Detailed ways
现在参考附图介绍本发明的示例性实施方式，然而，本发明可以用许多不同的形式来实施，并且不局限于此处描述的实施例，提供这些实施例是为了详尽地且完全地公开本发明，并且向所属技术领域的技术人员充分传达本发明的范围。对于表示在附图中的示例性实施方式中的术语并不是对本发明的限定。在附图中，相同的单元/元件使用相同的附图标记。Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units/elements are marked with the same reference numerals.
除非另有说明，此处使用的术语(包括科技术语)对所属技术领域的技术人员具有通常的理解含义。另外，可以理解的是，以通常使用的词典限定的术语，应当被理解为与其相关领域的语境具有一致的含义，而不应该被理解为理想化的或过于正式的意义。Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.
图1为根据本发明实施方式的继电保护缺陷数据的唯一性校核方法100的流程图。如图1所示，本发明实施方式提供的继电保护缺陷数据的唯一性校核方法，计算待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度；根据多个样本缺陷记录对，利用支持向量机确定缺陷记录唯一性校核分界面；根据所述待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度对应的数据点与所述缺陷记录唯一性校核分界面的相对位置，确定所述待校验的缺陷记录对的唯一性校核结果；本发明的方法能够有效支撑继电保护专业管理人员分析继电保护缺陷，能够准确地判断同一设备的任意2次缺陷记录是否重复，能够减轻人工分析数据的工作量，减小了人力物力成本，通过筛查、清理重复的缺陷数据，能够提升继电保护缺陷数据的综合应用水平。本发明实施方式提供的继电保护缺陷数据的唯一性校核方法100，从步骤101处开始，在步骤101获取同一继电保护设备标识对应的至少两条待校验的缺陷记录，并选取所述至少两条待校验的缺陷记录中的两条待校验的缺陷记录作为待校验的缺陷记录对。FIG1 is a flow chart of a uniqueness verification method 100 for relay protection defect data according to an embodiment of the present invention. As shown in FIG1 , the uniqueness verification method for relay protection defect data provided by the embodiment of the present invention calculates the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be verified; determines the defect record uniqueness verification interface using a support vector machine based on multiple sample defect record pairs; determines the uniqueness verification result of the defect record pair to be verified based on the relative position of the data points corresponding to the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be verified and the defect record uniqueness verification interface; the method of the present invention can effectively support professional managers of relay protection to analyze relay protection defects, can accurately determine whether any two defect records of the same equipment are repeated, can reduce the workload of manual data analysis, reduce the cost of manpower and material resources, and can improve the comprehensive application level of relay protection defect data by screening and cleaning up duplicate defect data. The uniqueness verification method 100 of relay protection defect data provided in an embodiment of the present invention starts from step 101. In step 101, at least two defect records to be verified corresponding to the same relay protection device identifier are obtained, and two defect records to be verified from the at least two defect records to be verified are selected as defect record pairs to be verified.
继电保护设备的实物标识I D是继电保护设备的唯一性编码，是在不同分析、应用、管理场合下唯一地表征一台保护设备的编码规范。因此，本发明据此建立继电保护缺陷数据索引并抽取缺陷记录对来校验缺陷数据的唯一性。The physical identification ID of the relay protection device is the unique code of the relay protection device, and is a coding specification that uniquely represents a protection device in different analysis, application, and management situations. Therefore, the present invention establishes a relay protection defect data index and extracts defect record pairs to verify the uniqueness of the defect data.
在本发明的实施方式中，继电保护缺陷数据唯一性校核的过程是：首先，以继电保护设备的实物标识ID为查询条件，在继电保护缺陷数据表中查找相同的实物标识ID对应的多条缺陷记录，并根据返回的查询结果信息随机抽取同一个实物标识ID对应的两条缺陷记录进行组合，获取待校核的缺陷记录对。然后，计算所述待检测的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度，根据待检测的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度判断是否为重复记录，并将重复记录返回数据源端处理。In the implementation mode of the present invention, the process of checking the uniqueness of relay protection defect data is as follows: first, using the physical identification ID of the relay protection device as the query condition, searching for multiple defect records corresponding to the same physical identification ID in the relay protection defect data table, and randomly extracting two defect records corresponding to the same physical identification ID according to the returned query result information to combine them, and obtaining the defect record pair to be checked. Then, the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be detected are calculated, and whether it is a duplicate record is determined according to the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be detected, and the duplicate record is returned to the data source for processing.
在步骤102，根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，并根据所述待校验的缺陷记录对中的缺陷属性数据以及每项缺陷属性的权重计算所述待校验的缺陷记录对的缺陷属性差异度。In step 102, the weight of each defect attribute is calculated based on the statistical results of various attributes of all defect records of relay protection, and the defect attribute difference of the defect record pair to be verified is calculated based on the defect attribute data in the defect record pair to be verified and the weight of each defect attribute.
优选地，其中所述方法利用如下方式根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，包括：Preferably, the method calculates the weight of each defect attribute according to the statistical results of various attributes of all defect records of relay protection in the following manner, including:
其中，ωi为第i项缺陷属性的权重；Pij为第i项缺陷属性的第j个取值出现的概率；k为第i项缺陷属性的不同取值的个数。Among them,ωi is the weight of the i-th defect attribute;Pij is the probability of the j-th value of the i-th defect attribute occurring; and k is the number of different values of the i-th defect attribute.
优选地，其中所述方法利用如下公式根据每项缺陷属性的权重计算所述缺陷记录对的缺陷属性差异度，包括：Preferably, the method calculates the defect attribute difference of the defect record pair according to the weight of each defect attribute using the following formula, including:
其中，V1为缺陷记录对的缺陷属性差异度；ωi为第i项缺陷属性的权重；Ii表示缺陷记录对第i项缺陷属性的数据是否有差别，如果有差别则取1，无差别则取0。Wherein,V1 is the difference degree of defect attribute of the defect record pair;ωi is the weight of the i-th defect attribute;Ii indicates whether the data of the defect record pair i-th defect attribute are different, if there is a difference, it takes 1, if there is no difference, it takes 0.
本发明方法的核心是分析重复记录与不重复记录在特征上的典型差异，确定判断缺陷记录对重复性的指标，确定分类原则，并据此判断缺陷记录对的重复情况。The core of the method of the present invention is to analyze the typical differences in features between duplicate records and non-duplicate records, determine indicators for judging the repeatability of defective record pairs, determine classification principles, and judge the repeatability of defective record pairs accordingly.
具有相同实物标识ID的两条缺陷记录若为不同的记录，那么，这2条记录在内容上应当有一定差异。例如，对于继电保护缺陷数据，不同的缺陷在缺陷程度、缺陷设备分类、缺陷原因、保护是否退出、缺陷部位、具体缺陷原因上应当有一定差别，那么差异度指标可以通过反映待校验缺陷在这些属性上是否存在差异、差异程度来构建，因此设置了缺陷属性差异度指标。另外，对于不同的记录，两条记录在时间上的差异性是唯一性校核的关键特征。继电保护缺陷数据中，与时间相关的属性包括缺陷发生时间、缺陷消除时间、保护退出时间和保护恢复时间。由于缺陷并不一定导致保护退出运行，保护退出时间、保护恢复时间并不一定存在，所以基于缺陷发生时间、缺陷消除时间的差异，同时设置了将缺陷时间差异度作为缺陷记录差异度的校核指标。除了缺陷发生时间、缺陷消除时间的差异度外，由于继电保护缺陷数据是一种运维数据，现场填报具有一定主观性，由于运维人员填报的缺陷发生和消除时间为预估时间，重复填报的缺陷数据在缺陷发生时间及缺陷消除时间具备一定差异，导致单独使用时间绝对值作为指标的判断方法将此类重复数据误判为非重复数据。然而，现场对于相同缺陷填报的多次记录在缺陷时间段上通常相交，依据相交程度增设指标对于此类记录的唯一性校核的针对性强。因此，同时将缺陷时间跨度分隔度作为校核指标。If two defect records with the same physical identification ID are different records, then the two records should have certain differences in content. For example, for relay protection defect data, different defects should have certain differences in defect degree, defect equipment classification, defect cause, whether protection is exited, defect location, and specific defect cause. Then the difference index can be constructed by reflecting whether there are differences and the degree of difference in these attributes of the defects to be verified, so a defect attribute difference index is set. In addition, for different records, the difference in time between the two records is a key feature for uniqueness verification. In relay protection defect data, time-related attributes include defect occurrence time, defect elimination time, protection exit time, and protection recovery time. Since defects do not necessarily cause protection to exit operation, protection exit time and protection recovery time do not necessarily exist, so based on the difference in defect occurrence time and defect elimination time, the defect time difference is set as a verification indicator for the defect record difference. In addition to the difference between the defect occurrence time and the defect elimination time, since the relay protection defect data is a kind of operation and maintenance data, the on-site reporting is subjective. Since the defect occurrence and elimination time reported by the operation and maintenance personnel are estimated times, the defect data reported repeatedly have certain differences in the defect occurrence time and the defect elimination time, resulting in the judgment method that uses the absolute value of time as the indicator alone to misjudge such repeated data as non-repeated data. However, multiple records of the same defect report on site usually intersect in the defect time period. Adding indicators based on the degree of intersection is highly targeted for the uniqueness verification of such records. Therefore, the defect time span separation degree is also used as a verification indicator.
在本发明的实施方式中，根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，并根据所述待校验的缺陷记录对中的缺陷属性数据以及每项缺陷属性的权重计算所述待校验的缺陷记录对的缺陷属性差异。In an embodiment of the present invention, the weight of each defect attribute is calculated based on the statistical results of various attributes of all defect records of relay protection, and the defect attribute difference of the defect record pair to be verified is calculated based on the defect attribute data in the defect record pair to be verified and the weight of each defect attribute.
继电保护设备的缺陷属性数据主要体现在缺陷程度、缺陷设备分类、缺陷原因、保护是否退出、缺陷部位和具体缺陷原因中，缺陷属性的差异度指标定义为两条记录间不同属性的数目占总数目的比重。上述缺陷属性包含的信息量有一定差别，例如：保护是否退出仅有2个选项，缺陷部位有64个选项，显然，缺陷部位属性更能反映出两条缺陷的差异度，在缺陷信息的差异度上，应有更高的权重。The defect attribute data of relay protection equipment are mainly reflected in the degree of defect, classification of defective equipment, cause of defect, whether protection is exited, defect location and specific cause of defect. The difference index of defect attribute is defined as the proportion of the number of different attributes between two records to the total number. The amount of information contained in the above defect attributes is different. For example, there are only 2 options for whether protection is exited, and 64 options for defect location. Obviously, the defect location attribute can better reflect the difference between the two defects, and should have a higher weight in the difference of defect information.
信息熵是度量样本集合纯度最常用的一种指标。假定当前样本集合D中第k类样本所占的比例为pk(k＝1，2,…,|y|)，则D的信息熵定义为：Ent(D)的值越小，则D的纯度越高。Information entropy is the most commonly used indicator to measure the purity of a sample set. Assuming that the proportion of the k-th class of samples in the current sample set D is pk (k = 1, 2, ..., |y|), the information entropy of D is defined as: The smaller the value of Ent(D), the higher the purity of D.
因此，本发明中不同缺陷属性的权重依据熵理论设置。利用如下方式根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，包括：Therefore, the weights of different defect attributes in the present invention are set according to the entropy theory. The weight of each defect attribute is calculated according to the statistical results of the attributes of all defect records of the relay protection in the following manner, including:
其中，ωi为第i项缺陷属性的权重；Pij为第i项缺陷属性的第j个取值出现的概率；k为第i项缺陷属性的不同取值的个数。Among them,ωi is the weight of the i-th defect attribute;Pij is the probability of the j-th value of the i-th defect attribute occurring; and k is the number of different values of the i-th defect attribute.
例如，计算的不同缺陷属性的权重如表1所示。For example, the weights of different defect attributes are calculated as shown in Table 1.
表1不同缺陷属性的权重设置Table 1 Weight settings for different defect attributes
利用如下公式根据每项缺陷属性的权重计算所述缺陷记录对的缺陷属性差异度，包括：The defect attribute difference of the defect record pair is calculated according to the weight of each defect attribute using the following formula, including:
其中，V1为缺陷记录对的缺陷属性差异度；ωi为第i项缺陷属性的权重；Ii表示缺陷记录对第i项缺陷属性的数据是否有差别，如果有差别则取1，无差别则取0。若缺陷属性为表1中的6项，则i为6。Among them,V1 is the difference degree of defect attribute of the defect record pair;ωi is the weight of the i-th defect attribute;Ii indicates whether the defect record has a difference in the data of the i-th defect attribute. If there is a difference, it takes 1, and if there is no difference, it takes 0. If the defect attribute is the 6 items in Table 1, then i is 6.
在步骤103，根据所述待校验的缺陷记录对中的缺陷时间确定所述待校验的缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度。In step 103, the defect time difference and defect time span separation of the defect record pair to be verified are determined according to the defect time in the defect record pair to be verified.
优选地，其中所述方法利用如下方式根据缺陷记录对中的缺陷时间确定所述缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度，包括：Preferably, the method determines the defect time difference and defect time span separation of the defect record pair according to the defect time in the defect record pair in the following manner, including:
V2＝(V21+V22)/2，V2 = (V21 +V22 ) / 2,
其中，V2为缺陷记录对的缺陷时间差异度；V21为缺陷记录对的缺陷发生时间差异度；V22为缺陷记录对的缺陷消除时间差异度；Tdif1和Tdif2分别为缺陷记录对中的缺陷发生时间之差和缺陷消除时间之差，单位为分钟；V3为缺陷记录对的缺陷时间跨度分隔度。Among them,V2 is the defect time difference of the defect record pair;V21 is the defect occurrence time difference of the defect record pair;V22 is the defect elimination time difference of the defect record pair;Tdif1 andTdif2 are the difference in defect occurrence time and defect elimination time in the defect record pair, respectively, in minutes;V3 is the defect time span separation of the defect record pair.
由于不同缺陷记录的时间差分布在分钟到年范围内，缺陷时间的差异度指标对时间差进行对数计算，将具有不同量级时间差的校验对象的指标值合理分布在(0,1)区间上。对于某年度继电保护缺陷数据唯一性校核的应用场景，相同继电保护实物标识ID的缺陷记录时间差不超过1年，缺陷记录对的缺陷时间差异度为：Since the time difference of different defect records ranges from minutes to years, the defect time difference index calculates the logarithm of the time difference, and reasonably distributes the index values of the verification objects with different magnitudes of time difference in the interval (0,1). For the application scenario of uniqueness verification of relay protection defect data in a certain year, the time difference of defect records with the same relay protection physical identification ID does not exceed 1 year, and the defect time difference of the defect record pair is:
V2＝(V21+V22)/2，V2 = (V21 +V22 ) / 2,
其中，V2为缺陷记录对的缺陷时间差异度；V21为缺陷记录对的缺陷发生时间差异度；V22为缺陷记录对的缺陷消除时间差异度；Tdif1和Tdif2分别为缺陷记录对中的缺陷发生时间之差和缺陷消除时间之差，单位为分钟。Among them, V2 is the defect time difference of the defect record pair; V21 is the defect occurrence time difference of the defect record pair; V22 is the defect elimination time difference of the defect record pair; Tdif1 and Tdif2 are the difference in defect occurrence time and defect elimination time in the defect record pair, respectively, in minutes.
按照上述公式，计算得到不同的缺陷差异时间的差异度指标值如表2所示。According to the above formula, the difference index values of different defect difference times are calculated as shown in Table 2.
表2不同缺陷差异时间下的差异度指标值Table 2 Difference index values under different defect difference time
在本发明的实施方式中，利用如下公式计算缺陷记录对的缺陷时间跨度分隔度V3，包括：In the embodiment of the present invention, the defect time span separation V3 of the defect record pair is calculated using the following formula, including:
如图2所示，缺陷时间跨度交集与并集的计算方法是提取缺陷记录对中的2条缺陷记录的发生及消除时间构成缺陷持续时间区间，2个区间的交集、并集便是缺陷时间跨度的交集、并集。As shown in Figure 2, the calculation method of the intersection and union of defect time spans is to extract the occurrence and elimination time of the two defect records in the defect record pair to form the defect duration interval. The intersection and union of the two intervals are the intersection and union of the defect time spans.
在步骤104，获取多个样本缺陷记录对，计算每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度，并根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面。In step 104, a plurality of sample defect record pairs are obtained, and the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair are calculated. Based on the data points consisting of the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair, a support vector machine is used to determine the defect record uniqueness verification interface.
优选地，其中所述根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面，包括：Preferably, the method of determining the defect record uniqueness verification interface using a support vector machine based on the data points consisting of the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair includes:
构建以缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度为坐标的指标空间；Construct an index space with defect attribute difference, defect time difference and defect time span separation as coordinates;
在所述指标空间中确定每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，并根据每个样本缺陷记录对的是否为重复缺陷记录的先验结果进行分簇，确定重复缺陷记录簇和非重复缺陷记录簇；Determine the data points consisting of defect attribute difference, defect time difference and defect time span separation of each sample defect record pair in the index space, and perform clustering according to the prior result of whether each sample defect record pair is a duplicate defect record to determine a duplicate defect record cluster and a non-duplicate defect record cluster;
根据所述重复缺陷簇和非重复缺陷簇，采用支持向量机进行训练，以确定缺陷记录唯一性校核分界面。According to the repeated defect clusters and the non-repeated defect clusters, a support vector machine is used for training to determine a defect record uniqueness verification interface.
支持向量机(Support Vector Machine,SVM)是一种性能良好的分类工具，其基本思想是在样本空间或特征空间构造出最优超平面，使得超平面与不同类样本集之间的距离最大，从而达到最大的泛化能力。SVM是从线性可分情况下的最优分类面发展而来的，基本思想可用图3的两类线性可分情况说明。实心点和空心点代表两类样本，实线P0、P1为分类线。Support Vector Machine (SVM) is a classification tool with good performance. Its basic idea is to construct an optimal hyperplane in the sample space or feature space so that the distance between the hyperplane and the sample sets of different classes is maximized, thereby achieving the maximum generalization ability. SVM is developed from the optimal classification surface in the case of linear separability. The basic idea can be explained by the two types of linear separability in Figure 3. The solid points and hollow points represent two types of samples, and the solid lines P0 and P1 are classification lines.
设线性可分样本集为：Assume that the linearly separable sample set is:
(x1,y1),(x2,y2),…,(xN,yN)；xi∈Rd；yi∈{1,-1} (1)(x1 ,y1 ),(x2 ,y2 ),…,(xN ,yN );xi ∈Rd ;yi ∈{1,-1} (1)
其中，x为d维向量，y为类别标号，即将样本分为两类。Among them, x is a d-dimensional vector and y is the category label, which means that the samples are divided into two categories.
存在超平面公式如下：The hyperplane formula is as follows:
f(x)＝(w·x)-b＝0 (2)f(x)＝(w·x)-b＝0 (2)
可以将所有N个样本无错误分开。其中w是d维权值相量；b为常数项。All N samples can be separated without error. Where w is the d-dimensional weight phasor; b is a constant term.
若存在一个超平面，可以使训练样本没有错误的分开，并且各类训练样本中离超平面最近的样本与超平面之间的距离尽可能的大，则该超平面被称为最优分类超平面。寻找该最优超平面的过程即为支持向量机的模型训练过程。If there is a hyperplane that can separate the training samples without error, and the distance between the sample closest to the hyperplane and the hyperplane is as large as possible, then the hyperplane is called the optimal classification hyperplane. The process of finding the optimal hyperplane is the model training process of the support vector machine.
该算法的优点是不会过多的受到噪声数据的影响，而且不容易出现过拟合，较之神经网络算法更容易使用；其缺点是寻找最好的模型需要测试不同的和函数和模型参数组合，训练缓慢，形成的复杂的黑箱模型难以解释。支持向量机算法主要应用于模式识别、函数逼近和概率密度估计等领域。The advantage of this algorithm is that it is not overly affected by noisy data, and it is not prone to overfitting, making it easier to use than neural network algorithms; its disadvantage is that finding the best model requires testing different combinations of functions and model parameters, which is slow to train and forms a complex black box model that is difficult to explain. Support vector machine algorithms are mainly used in pattern recognition, function approximation, and probability density estimation.
在本发明的实施方式中，在继电保护实物标识I D相同的多条记录中，选取任意2条记录，确定多个缺陷记录样本对，并计算上述三个缺陷差异度指标。由于已知缺陷记录是否为重复记录，因此在空间上进行描点，能够得到重复缺陷的点簇和非重复缺陷的点簇，通过利用SVM分析训练，即可得到缺陷记录唯一性校核分界面。对于待校核的缺陷记录对，计算缺陷记录对的三个指标，利用缺陷记录唯一性校核分界面即可判别是否为重复缺陷记录。In the embodiment of the present invention, any two records are selected from multiple records with the same relay protection physical identification ID, multiple defect record sample pairs are determined, and the above three defect difference indices are calculated. Since it is known whether the defect record is a duplicate record, the point clusters of duplicate defects and non-duplicate defects can be obtained by plotting points in space, and the defect record uniqueness verification interface can be obtained by using SVM analysis training. For the defect record pair to be verified, the three indicators of the defect record pair are calculated, and the defect record uniqueness verification interface can be used to determine whether it is a duplicate defect record.
例如，以2019年继电保护数据为分析对象。以保护设备实物标识ID为关键属性，建立继电保护缺陷索引，如：{1995305,{0500000000002190,0500000000002083,0500000000002075}}。其中，1995305为保护设备实物标识ID，0500000000002190等3个实物标识ID为1995305的保护设备的3条缺陷记录的编号。将该保护设备的3条缺陷记录任取2条进行排列组合，得到3个缺陷对。For example, the relay protection data of 2019 is taken as the analysis object. The relay protection defect index is established with the physical identification ID of the protection equipment as the key attribute, such as: {1995305,{0500000000002190,0500000000002083,0500000000002075}}. Among them, 1995305 is the physical identification ID of the protection equipment, and 0500000000002190 and other three defect records of the protection equipment with the physical identification ID of 1995305 are numbers. Randomly select two of the three defect records of the protection equipment and arrange and combine them to obtain three defect pairs.
继电保护缺陷唯一性校核模型是两类线性可分问题，继电保护缺陷唯一性校核的输入是3个指标—缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度，输出为缺陷对是否是重复的，即重复缺陷记录对和非重复缺陷记录对。取2019年所有缺陷数据作为训练样本，共收集了737个缺陷对，来自624台保护设备。计算737个缺陷记录对的缺陷属性差异度V1、缺陷时间差异度V2、缺陷时间跨度分隔度V3三项指标。为了得到数据唯一性校核规则，需人为判断这737个缺陷对是否为重复缺陷。再以V1、V2、V3为坐标构建指标空间，即构建以V1、V2、V3三个指标作为横轴、竖轴、纵轴坐标的3维直角坐标体系，那么空间中任一点是一个以缺陷记录对的三项指标为坐标的空间点。结合其实际是否为重复记录的人为判断，对重复缺陷记录对及非重复缺陷记录对分别描绘，得到点簇。在737个缺陷记录对中，有512个记录对是不重复的，其点簇如图4圆点所示，剩余225个记录对是重复的，其点簇用如图4三角形点所示。采用支持向量机能够训练得到最优分类面，方程为：The relay protection defect uniqueness verification model is a two-type linear separable problem. The input of the relay protection defect uniqueness verification is three indicators—defect attribute difference, defect time difference, and defect time span separation. The output is whether the defect pair is repeated, that is, repeated defect record pairs and non-repeated defect record pairs. All defect data in 2019 were taken as training samples, and a total of 737 defect pairs were collected from 624 protection devices. The three indicators of defect attribute difference V1 , defect time difference V2 , and defect time span separation V3 of the 737 defect record pairs were calculated. In order to obtain the data uniqueness verification rules, it is necessary to manually judge whether these 737 defect pairs are repeated defects. Then construct the indicator space with V1 , V2 , and V3 as coordinates, that is, construct a 3D rectangular coordinate system with V1 , V2 , and V3 as the horizontal, vertical, and ordinate coordinates. Then any point in the space is a spatial point with the three indicators of the defect record pair as coordinates. Combined with the human judgment of whether it is a duplicate record, the duplicate defect record pairs and non-duplicate defect record pairs are depicted separately to obtain point clusters. Among the 737 defect record pairs, 512 record pairs are non-duplicate, and their point clusters are shown as circles in Figure 4. The remaining 225 record pairs are duplicates, and their point clusters are shown as triangles in Figure 4. The support vector machine can be trained to obtain the optimal classification surface, and the equation is:
0.0096V1+0.0504V2+0.982V3-1＝00.0096V1 +0.0504V2 +0.982V3 -1＝0
该最优平面即图4中空间平面。V1、V2、V3分别表示缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度指标。本式中，与公式(1)(2)对应地，x＝[V1,V2,V3]，w＝[0.0096,0.0504,0.982],b＝1。The optimal plane is the spatial plane in Figure 4. V1 , V2 , and V3 represent the defect attribute difference, defect time difference, and defect time span separation indexes, respectively. In this formula, corresponding to formulas (1) and (2), x = [V1 , V2 , V3 ], w = [0.0096, 0.0504, 0.982], and b = 1.
在步骤105，根据所述待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度对应的数据点与所述缺陷记录唯一性校核分界面的相对位置，确定所述待校验的缺陷记录对的唯一性校核结果；其中，所述唯一性校核结果包括：重复缺陷记录对和非重复缺陷记录对。In step 105, the uniqueness verification result of the defect record pair to be verified is determined according to the relative position of the data points corresponding to the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be verified and the defect record uniqueness verification interface; wherein the uniqueness verification result includes: duplicate defect record pairs and non-duplicate defect record pairs.
具体地，对于一组新的缺陷记录对，计算该缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度指标，构成[V1,V2,V3]，代入方程f(V1,V2,V3)＝0.0096V1+0.0504V2+0.982V3-1。若f(V1,V2,V3)≥0，表示该缺陷对是非重复的，若f(V1,V2,V3)<0，表示该缺陷对是重复的。Specifically, for a new set of defect record pairs, the defect attribute difference, defect time difference and defect time span separation index of the defect record pair are calculated to form [V1 ,V2 ,V3 ], and substituted into the equation f(V1 ,V2 ,V3 )＝0.0096V1 +0.0504V2 +0.982V3 -1. If f(V1 ,V2 ,V3 )≥0, it means that the defect pair is non-repetitive, and if f(V1 ,V2 ,V3 )<0, it means that the defect pair is repetitive.
在本发明的实施方式中，在模型校验阶段，选取2020年保护设备缺陷数据，依据继电保护设备的实物标识ID构建缺陷索引，在索引内部选取相同保护实物标识ID的缺陷记录对，计算三项指标(V1,V2,V3)，判断该空间点相对分界面的位置，从而将该缺陷记录对划分到重复缺陷记录或非重复缺陷记录区域中，判断缺陷记录对是否为重复缺陷记录，实现继电保护缺陷数据的唯一性校核。In an embodiment of the present invention, in the model verification stage, the protection equipment defect data in 2020 is selected, and a defect index is constructed based on the physical identification ID of the relay protection equipment. The defect record pairs with the same protection physical identification ID are selected within the index, and three indicators (V1 ,V2 ,V3 ) are calculated to determine the position of the spatial point relative to the interface, thereby dividing the defect record pair into a duplicate defect record or a non-duplicate defect record area, and determining whether the defect record pair is a duplicate defect record, thereby realizing the uniqueness verification of the relay protection defect data.
在缺陷唯一性校核的3项指标中，缺陷时间跨度分隔度指标V3发挥了良好的分类作用。不重复缺陷的V3靠近1，重复缺陷的V3小于1。此外，缺陷时间的差异度指标V2也有良好的分类效果。通常，缺陷时间差异度>0.5则为不重复缺陷，对应缺陷对中2个缺陷的发生时间及缺陷消除时间的时间差为1日以上。而缺陷属性的差异度指标的分类效果最差，原因是即使是相同缺陷，现场不同运行人员对该缺陷的理解有一定差距，导致反映相同缺陷的两条缺陷记录在缺陷部位、具体缺陷原因等信息的填报上不一致。Among the three indicators for defect uniqueness verification, the defect time span separation index V3 played a good classification role. V3 of non-repeating defects is close to 1, and V3 of repeating defects is less than 1. In addition, the defect time difference index V2 also has a good classification effect. Generally, defect time difference > 0.5 is a non-repeating defect, and the time difference between the occurrence time of the two defects in the corresponding defect pair and the defect elimination time is more than 1 day. The defect attribute difference index has the worst classification effect. The reason is that even if it is the same defect, there is a certain gap in the understanding of the defect by different on-site operators, resulting in inconsistency in the reporting of information such as the defect location and specific defect cause of the two defect records reflecting the same defect.
本发明的方法能够有效支撑重复继电保护缺陷数据筛查，给出相同装置的任意2次缺陷是否为重复记录的结论，判断准确率高。将疑似重复的缺陷记录提供给专业人员，专业人员仅需少量投入即可确定是否为重复录入的缺陷，避免专业人员直接对海量继电保护缺陷数据进行唯一性判别及数据治理，减轻人工分析数据的工作量，有效支撑筛查、清理重复的缺陷数据，提升继电保护缺陷数据的综合应用水平。The method of the present invention can effectively support the screening of repeated relay protection defect data, and give a conclusion on whether any two defects of the same device are repeated records, with high judgment accuracy. Suspected repeated defect records are provided to professionals, who only need a small amount of investment to determine whether they are repeated defects, avoiding professionals from directly performing uniqueness judgment and data management on massive relay protection defect data, reducing the workload of manual data analysis, effectively supporting the screening and cleaning of repeated defect data, and improving the comprehensive application level of relay protection defect data.
本发明的方法在典型缺陷唯一性校核场景中的应用效果如下：The application effects of the method of the present invention in typical defect uniqueness verification scenarios are as follows:
场景1：本方法针对设备的实物标识ID建立缺陷索引。对于不同保护设备的缺陷，缺陷不在相同索引下，本发明的方法能够自动判定为非重复缺陷数据；Scenario 1: This method establishes a defect index for the physical identification ID of the device. For defects of different protection devices, the defects are not under the same index, and the method of the present invention can automatically determine them as non-repetitive defect data;
场景2：相同保护设备的2条缺陷记录，从缺陷发生时间到缺陷消除时间的跨度区间没有交集，且2条记录的缺陷发生时间、缺陷消除时间有一定差距，采用本发明的方法能够判定2条缺陷记录为非重复记录，从而有效减少人员人工分析的工作量；Scenario 2: There are two defect records of the same protection device, and the span interval from the defect occurrence time to the defect elimination time has no intersection, and there is a certain gap between the defect occurrence time and the defect elimination time of the two records. The method of the present invention can determine that the two defect records are non-duplicate records, thereby effectively reducing the workload of manual analysis by personnel;
场景3：相同缺陷重复填报，且填报的离散型项目(具体是缺陷程度、缺陷设备分类、缺陷原因等)差异度大。由于2条缺陷记录的缺陷发生及消除时间相近，本发明能够排除离散型项目差异度大的干扰，准确判断为重复缺陷。例如，2020年发生的一起因电源缺陷引起通道异常的缺陷，运维人员1填报为通道接口设备缺陷、由装置制造不良引起、一般缺陷、责任在制造部门、缺陷部位在电源插件；运维人员2填报为通信传输设备缺陷、由其他原因引起、严重缺陷、责任在其他部门、缺陷部位在光缆(OPGW或ADSS)。该缺陷对的[V1,V2,V3]＝[1,0,0]。事实上，指标V1对于判别设备是否为重复缺陷的作用效果有限，如图4中分界面对该指标的斜率极低，该指标仅能起到辅助判断的作用；Scenario 3: The same defect is reported repeatedly, and the reported discrete items (specifically, defect degree, defect equipment classification, defect cause, etc.) are very different. Since the defect occurrence and elimination time of the two defect records are similar, the present invention can eliminate the interference of large differences in discrete items and accurately judge it as a repeated defect. For example, in 2020, a defect caused by a power supply defect caused a channel abnormality. Operation and maintenance personnel 1 reported it as a channel interface equipment defect, caused by poor manufacturing of the device, a general defect, the responsibility lies with the manufacturing department, and the defect is located in the power plug; Operation and maintenance personnel 2 reported it as a communication transmission equipment defect, caused by other reasons, a serious defect, the responsibility lies with other departments, and the defect is located in the optical cable (OPGW or ADSS). [V1 ,V2 ,V3 ] = [1, 0, 0] for this defect pair. In fact, the indicatorV1 has limited effect on determining whether the equipment is a repeated defect. As shown in Figure 4, the slope of the interface to this indicator is extremely low, and this indicator can only serve as an auxiliary judgment;
场景4：同一台装置先后发生2次相同原因的缺陷并采用相同方法消缺。由于这2次缺陷发生到消除的时间跨度无重合且发生时间、消除时间有一定差异，应用本方法能够正确判断出2次缺陷非重复性缺陷。例如，2020年的消缺记录中，更换某设备损坏的电源插件3天后，电源插件再次损坏、更换。该缺陷对的[V1,V2,V3]＝[0.1275,0.6167,1]；Scenario 4: The same device has two defects with the same cause and the same method is used to eliminate the defects. Since the time span from the occurrence to the elimination of these two defects does not overlap and there is a certain difference between the occurrence time and the elimination time, this method can correctly determine that the two defects are non-repetitive defects. For example, in the defect elimination record in 2020, 3 days after the damaged power plug of a certain device was replaced, the power plug was damaged again and replaced. The defect pair [V1 ,V2 ,V3 ] = [0.1275,0.6167,1];
场景5：同一台装置采用不同消缺方案均暂时消除了缺陷，但经一段不长的时间该装置发生其他缺陷，需要采用其他消缺手段消除缺陷。同场景4，由于2次缺陷发生到消除的时间跨度无重合且发生时间、消除时间有一定差异，本发明同样能够正确判断出2次缺陷非重复缺陷。例如，2020年的消缺记录中，某装置在更换电源插件后短暂运行一段时间又发生缺陷，随后更换CPU插件才再次完成消缺。该缺陷对的[V1,V2,V3]＝[0,0.5756,1.0000]；Scenario 5: The same device temporarily eliminates defects using different defect elimination schemes, but after a short period of time, other defects occur in the device and other defect elimination methods are needed to eliminate the defects. Similar to Scenario 4, since the time span from the occurrence to the elimination of the two defects does not overlap and there is a certain difference between the occurrence time and the elimination time, the present invention can also correctly determine that the two defects are non-repeating defects. For example, in the defect elimination record of 2020, a device had defects again after a short period of operation after the power plug-in was replaced, and then the CPU plug-in was replaced to complete the defect elimination again. The defect pair [V1 ,V2 ,V3 ] = [0,0.5756,1.0000];
场景6：同一台装置的两条缺陷记录的缺陷发生时间相近，而两条缺陷记录的缺陷消除时间中，其中一条记录的缺陷消除时间与缺陷发生时间相近，而另一条记录的缺陷消除时间与缺陷发生时间相距较长，导致缺陷时间的差异度V2较大。经分析，这2条缺陷记录是在缺陷发生到消除时间跨度上有交叉的缺陷，极有可能是重复缺陷，本发明设置的指标—缺陷时间跨度分隔度V3能够支撑此种场景下缺陷记录对的唯一性判别。图4表明，除非2条缺陷记录在缺陷发生到消除的时间区间内无交叉，否则有极大概率被划归到重复缺陷记录中去。2020年发生的一起CPU插件损坏的缺陷中，其中一条记录显示当日完成消缺，而另外一条记录显示2日后完成消缺，表明缺陷被重复填报。缺陷记录对的[V1,V2,V3]＝[0.0782,0.4952,0.8868]，本发明能够准确将此类案例划归到重复缺陷中。Scenario 6: The defect occurrence time of two defect records of the same device is similar, and in the defect elimination time of the two defect records, the defect elimination time of one record is close to the defect occurrence time, while the defect elimination time of the other record is far away from the defect occurrence time, resulting in a large difference in defect timeV2 . After analysis, these two defect records are defects that overlap in the time span from defect occurrence to elimination, and are very likely to be repeated defects. The indicator set by the present invention-the defect time span separationV3 can support the uniqueness judgment of defect record pairs in this scenario. Figure 4 shows that unless the two defect records do not overlap in the time interval from defect occurrence to elimination, there is a high probability that they will be classified as repeated defect records. In a defect of a damaged CPU plug-in that occurred in 2020, one record showed that the defect was eliminated on the same day, while the other record showed that the defect was eliminated 2 days later, indicating that the defect was reported repeatedly. [V1 , V2 , V3 ] = [0.0782, 0.4952, 0.8868] of the defect record pair. The present invention can accurately classify such cases as repeated defects.
利用本发明的方法，能够将疑似重复缺陷记录提交给专业人员，以便数据进一步处理。By utilizing the method of the present invention, suspected duplicate defect records can be submitted to professionals for further data processing.
本发明虽然仅以继电保护缺陷数据的唯一性校核为分析对象，但继电保护其他类型数据，如继电保护动作、巡检巡视、告警等类型数据在数据重复特性上与继电保护缺陷数据的重复特性有一定相似性，本发明所提方法能够在继电保护其他类型数据的唯一性校核上应用，提升继电保护数据质量，提升数据治理及综合应用水平。Although the present invention only takes the uniqueness verification of relay protection defect data as the analysis object, other types of relay protection data, such as relay protection action, inspection and patrol, alarm and other types of data have certain similarities in data repetition characteristics with the repetition characteristics of relay protection defect data. The method proposed in the present invention can be applied to the uniqueness verification of other types of relay protection data, improve the quality of relay protection data, and improve the level of data governance and comprehensive application.
图5为根据本发明实施方式的继电保护缺陷数据的唯一性校核系统500的结构示意图。如图5所示，本发明实施方式提供的继电保护缺陷数据的唯一性校核系统500，包括：待校验的缺陷记录对确定单元501、缺陷属性差异度确定单元502、缺陷时间差异度确定单元503、缺陷记录唯一性校核分界面确定单元504和唯一性校核结果确定单元505。Fig. 5 is a schematic diagram of the structure of a uniqueness checking system 500 for relay protection defect data according to an embodiment of the present invention. As shown in Fig. 5, the uniqueness checking system 500 for relay protection defect data provided by an embodiment of the present invention comprises: a defect record pair determination unit 501 to be verified, a defect attribute difference determination unit 502, a defect time difference determination unit 503, a defect record uniqueness checking interface determination unit 504 and a uniqueness checking result determination unit 505.
优选地，所述待校验的缺陷记录对确定单元501，用于获取同一继电保护设备标识对应的至少两条待校验的缺陷记录，并选取所述至少两条待校验的缺陷记录中的两条待校验的缺陷记录作为待校验的缺陷记录对。Preferably, the defect record pair determination unit 501 is used to obtain at least two defect records to be verified corresponding to the same relay protection device identifier, and select two defect records to be verified from the at least two defect records to be verified as the defect record pair to be verified.
优选地，所述缺陷属性差异度确定单元502，用于根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，并根据所述待校验的缺陷记录对中的缺陷属性数据以及每项缺陷属性的权重计算所述待校验的缺陷记录对的缺陷属性差异度。Preferably, the defect attribute difference determination unit 502 is used to calculate the weight of each defect attribute based on the statistical results of each attribute of the entire defect record of the relay protection, and calculate the defect attribute difference of the defect record pair to be verified based on the defect attribute data in the defect record pair to be verified and the weight of each defect attribute.
优选地，其中所述系统利用如下方式根据继电保护全体缺陷记录的各项属性的统计结果计算每项缺陷属性的权重，包括：Preferably, the system calculates the weight of each defect attribute according to the statistical results of various attributes of all defect records of relay protection in the following manner, including:
其中，ωi为第i项缺陷属性的权重；Pij为第i项缺陷属性的第j个取值出现的概率；k为第i项缺陷属性的不同取值的个数。Among them, ωi is the weight of the i-th defect attribute;Pij is the probability of the j-th value of the i-th defect attribute occurring; k is the number of different values of the i-th defect attribute.
优选地，其中所述系统利用如下公式根据每项缺陷属性的权重计算所述缺陷记录对的缺陷属性差异度，包括：Preferably, the system calculates the defect attribute difference of the defect record pair according to the weight of each defect attribute using the following formula, including:
其中，V1为缺陷记录对的缺陷属性差异度；ωi为第i项缺陷属性的权重；Ii表示缺陷记录对第i项缺陷属性的数据是否有差别，如果有差别则取1，无差别则取0。Wherein,V1 is the difference degree of defect attribute of the defect record pair;ωi is the weight of the i-th defect attribute;Ii indicates whether the data of the defect record pair i-th defect attribute are different, if there is a difference, it takes 1, if there is no difference, it takes 0.
优选地，所述缺陷时间差异度确定单元503，用于根据所述待校验的缺陷记录对中的缺陷时间确定所述待校验的缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度。Preferably, the defect time difference determination unit 503 is used to determine the defect time difference and defect time span separation of the defect record pair to be verified according to the defect time in the defect record pair to be verified.
优选地，其中所述系统利用如下方式根据缺陷记录对中的缺陷时间确定所述缺陷记录对的缺陷时间差异度和缺陷时间跨度分隔度，包括：Preferably, the system determines the defect time difference and defect time span separation of the defect record pair according to the defect time in the defect record pair in the following manner, including:
V2＝(V21+V22)/2，V2 = (V21 +V22 ) / 2,
其中，V2为缺陷记录对的缺陷时间差异度；V21为缺陷记录对的缺陷发生时间差异度；V22为缺陷记录对的缺陷消除时间差异度；Tdif1和Tdif2分别为缺陷记录对中的缺陷发生时间之差和缺陷消除时间之差，单位为分钟；V3为缺陷记录对的缺陷时间跨度分隔度。Among them,V2 is the defect time difference of the defect record pair;V21 is the defect occurrence time difference of the defect record pair;V22 is the defect elimination time difference of the defect record pair;Tdif1 andTdif2 are the difference in defect occurrence time and defect elimination time in the defect record pair, respectively, in minutes;V3 is the defect time span separation of the defect record pair.
优选地，所述缺陷记录唯一性校核分界面确定单元504，用于获取多个样本缺陷记录对，计算每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度，并根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面。Preferably, the defect record uniqueness verification interface determination unit 504 is used to obtain multiple sample defect record pairs, calculate the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair, and determine the defect record uniqueness verification interface using a support vector machine based on the data points consisting of the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair.
优选地，其中所述缺陷记录唯一性校核分界面确定单元504，根据每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，利用支持向量机确定缺陷记录唯一性校核分界面，包括：Preferably, the defect record uniqueness verification interface determination unit 504 determines the defect record uniqueness verification interface using a support vector machine according to the data points consisting of the defect attribute difference, defect time difference and defect time span separation of each sample defect record pair, including:
构建以缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度为坐标的指标空间；Construct an index space with defect attribute difference, defect time difference and defect time span separation as coordinates;
在所述指标空间中确定每个样本缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度组成的数据点，并根据每个样本缺陷记录对的是否为重复缺陷记录的先验结果进行分簇，确定重复缺陷记录簇和非重复缺陷记录簇；Determine the data points consisting of defect attribute difference, defect time difference and defect time span separation of each sample defect record pair in the index space, and perform clustering according to the prior result of whether each sample defect record pair is a duplicate defect record to determine a duplicate defect record cluster and a non-duplicate defect record cluster;
根据所述重复缺陷簇和非重复缺陷簇，采用支持向量机进行训练，以确定缺陷记录唯一性校核分界面。According to the repeated defect clusters and the non-repeated defect clusters, a support vector machine is used for training to determine a defect record uniqueness verification interface.
优选地，所述唯一性校核结果确定单元505，用于根据所述待校验的缺陷记录对的缺陷属性差异度、缺陷时间差异度和缺陷时间跨度分隔度对应的数据点与所述缺陷记录唯一性校核分界面的相对位置，确定所述待校验的缺陷记录对的唯一性校核结果；其中，所述唯一性校核结果包括：重复缺陷记录对和非重复缺陷记录对。Preferably, the uniqueness verification result determination unit 505 is used to determine the uniqueness verification result of the defect record pair to be verified based on the relative position of the data points corresponding to the defect attribute difference, defect time difference and defect time span separation of the defect record pair to be verified and the defect record uniqueness verification interface; wherein the uniqueness verification result includes: duplicate defect record pairs and non-duplicate defect record pairs.
本发明的实施例的继电保护缺陷数据的唯一性校核系统500与本发明的另一个实施例的继电保护缺陷数据的唯一性校核方法100相对应，在此不再赘述。The uniqueness checking system 500 of relay protection defect data in the embodiment of the present invention corresponds to the uniqueness checking method 100 of relay protection defect data in another embodiment of the present invention, which will not be described in detail here.
已经通过参考少量实施方式描述了本发明。然而，本领域技术人员所公知的，正如附带的专利权利要求所限定的，除了本发明以上公开的其他的实施例等同地落在本发明的范围内。The invention has been described above with reference to a few embodiments. However, it is readily apparent to a person skilled in the art that other embodiments than the ones disclosed above are equally within the scope of the invention, as defined by the appended patent claims.
通常地，在权利要求中使用的所有术语都根据他们在技术领域的通常含义被解释，除非在其中被另外明确地定义。所有的参考“一个/所述/该[装置、组件等]”都被开放地解释为所述装置、组件等中的至少一个实例，除非另外明确地说明。这里公开的任何方法的步骤都没必要以公开的准确的顺序运行，除非明确地说明。Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a/said/the [means, components, etc.]" are to be openly interpreted as at least one instance of said means, components, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not necessarily have to be performed in the exact order disclosed, unless explicitly stated otherwise.
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。The present application is described with reference to the flowchart and/or block diagram of the method, device (system) and computer program product according to the embodiment of the present application. It should be understood that each process and/or box in the flowchart and/or block diagram, and the combination of the process and/or box in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.
最后应当说明的是：以上实施例仅用以说明本发明的技术方案而非对其限制，尽管参照上述实施例对本发明进行了详细的说明，所属领域的普通技术人员应当理解：依然可以对本发明的具体实施方式进行修改或者等同替换，而未脱离本发明精神和范围的任何修改或者等同替换，其均应涵盖在本发明的权利要求保护范围之内。Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011478502.7ACN112732773B (en) | 2020-12-15 | 2020-12-15 | Method and system for checking uniqueness of relay protection defect data |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011478502.7ACN112732773B (en) | 2020-12-15 | 2020-12-15 | Method and system for checking uniqueness of relay protection defect data |
| Publication Number | Publication Date |
|---|---|
| CN112732773A CN112732773A (en) | 2021-04-30 |
| CN112732773Btrue CN112732773B (en) | 2024-04-16 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011478502.7AActiveCN112732773B (en) | 2020-12-15 | 2020-12-15 | Method and system for checking uniqueness of relay protection defect data |
| Country | Link |
|---|---|
| CN (1) | CN112732773B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113377759B (en)* | 2021-07-01 | 2023-05-16 | 贵州电网有限责任公司 | Defect filling data management method based on expert system algorithm |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930365A (en)* | 2012-09-24 | 2013-02-13 | 中国电力科学研究院 | Method for checking statistical analysis and operation management data of relay protection |
| CN102928720A (en)* | 2012-11-07 | 2013-02-13 | 广东电网公司 | Defect rate detecting method of oil immersed type main transformer |
| CN105427029A (en)* | 2015-11-05 | 2016-03-23 | 广州兴森快捷电路科技有限公司 | Verification system and method for defect detection |
| CN111222582A (en)* | 2020-01-13 | 2020-06-02 | 中国电力科学研究院有限公司 | Method and system for extracting defect mode of relay protection device |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5498258B2 (en)* | 2010-05-20 | 2014-05-21 | 株式会社東芝 | Defect analysis method, defect analysis apparatus, defect analysis system, and program |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930365A (en)* | 2012-09-24 | 2013-02-13 | 中国电力科学研究院 | Method for checking statistical analysis and operation management data of relay protection |
| CN102928720A (en)* | 2012-11-07 | 2013-02-13 | 广东电网公司 | Defect rate detecting method of oil immersed type main transformer |
| CN105427029A (en)* | 2015-11-05 | 2016-03-23 | 广州兴森快捷电路科技有限公司 | Verification system and method for defect detection |
| CN111222582A (en)* | 2020-01-13 | 2020-06-02 | 中国电力科学研究院有限公司 | Method and system for extracting defect mode of relay protection device |
| Title |
|---|
| Relay protection data integrity check method based on big data association algorithm;Guo Peng;2019 International Conference on Electronic Engineering and Informatics;20191231;506-508* |
| 基于模糊综合评估的 继电保护装置运行状态感知方法;杨国生 等;电力信息与通信技术;20200415;第18卷(第4期);74-83* |
| Publication number | Publication date |
|---|---|
| CN112732773A (en) | 2021-04-30 |
| Publication | Publication Date | Title |
|---|---|---|
| CN110824270B (en) | Electricity stealing user identification method and device combining transformer area line loss and abnormal events | |
| CN113344134B (en) | Low-voltage distribution monitoring terminal data acquisition abnormality detection method and system | |
| CN113189418B (en) | Topological relation identification method based on voltage data | |
| CN118094531B (en) | A real-time early warning integrated system for security operation and maintenance | |
| CN114201374A (en) | Operation and maintenance time sequence data anomaly detection method and system based on hybrid machine learning | |
| CN111639850A (en) | Quality evaluation method and system for multi-source heterogeneous data | |
| CN112418687B (en) | User electricity utilization abnormity identification method and device based on electricity utilization characteristics and storage medium | |
| CN117154716B (en) | Planning method and system for accessing distributed power supply into power distribution network | |
| CN120044302A (en) | Intelligent electric energy meter capable of monitoring and reporting operation errors | |
| CN118917840A (en) | Intelligent data optimization management platform based on electric power operation and maintenance | |
| CN112732773B (en) | Method and system for checking uniqueness of relay protection defect data | |
| CN120214678A (en) | A method and system for diagnosing electric meter faults based on big data analysis | |
| CN120177930A (en) | A method and system for sensing abnormal state of distribution network based on data processing | |
| CN118942606B (en) | Integrated intelligent integrated management and control system for hospital information | |
| CN118211882B (en) | Product quality management system and method based on big data | |
| CN117952318A (en) | Industrial garden carbon emission data management system and method based on big data | |
| CN114139956B (en) | Method and device for detecting line loss abnormality of transformer area | |
| CN115658772A (en) | Unmanned aerial vehicle photovoltaic inspection data asset management method and system | |
| CN114493120A (en) | Voltage data quality evaluation method and system and computer storage medium | |
| CN119624267B (en) | Power distribution network team evaluation index differentiation analysis system based on k-means clustering | |
| CN117436770B (en) | BIM-based building engineering construction management method and system | |
| CN118569685B (en) | Harbor big data-based production cost accounting method | |
| CN120030202B (en) | Data verification management method, system and medium based on artificial intelligence technology | |
| CN120331872B (en) | Tunnel construction safety monitoring system based on the Internet of Things | |
| CN118966813B (en) | A method, device, electronic device and storage medium for automatically locating high-loss transmission area |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |