技术领域Technical Field
本发明属于智能电网资产管理领域，尤其涉及一种基于案例驱动的电网资产全寿命管理方法及系统。The present invention belongs to the field of smart grid asset management, and in particular relates to a case-driven full-lifecycle management method and system for grid assets.
背景技术Background Art
现代智能电网在传统电力系统基础上融入新能源、电力电子技术、物理信息系统、需求侧响应等，使其成为具有高维、强非线性、动态过程复杂等特征的新型能源系统。为适应新一代电力系统广泛互联、智能互动、灵活柔性、安全可控的发展需要，电网资产快速增长，各资产智能量测数据呈几何级数增长。电网运行进入大数据和云计算时代，数据促进电网资产管理和电网运行业务进行深度融合。Modern smart grids integrate new energy, power electronics technology, physical information systems, demand-side response, etc. on the basis of traditional power systems, making them new energy systems with high-dimensional, strong nonlinear, and complex dynamic processes. In order to meet the development needs of the new generation of power systems with extensive interconnection, intelligent interaction, flexibility, safety and controllability, power grid assets are growing rapidly, and the intelligent measurement data of various assets is growing exponentially. Power grid operation has entered the era of big data and cloud computing, and data promotes the deep integration of power grid asset management and power grid operation business.
电力资产在其生命周期内发挥最大潜能对有效提高电网公司经济效益、电力系统安全稳定运行起到至关重要的作用。目前，电网设备管理都通过被动的计划检修完成，忽略对设备健康状况的评估与管理。随着电网资产量测数据、设备监控数据、各类管理系统数据呈几何级增长，电网公司需要在设备全寿命周期内使用数据驱动检修业务计划以及资产置换，以确保系统安全、可靠和经济运行。Maximizing the potential of power assets during their life cycle plays a vital role in effectively improving the economic benefits of power grid companies and the safe and stable operation of power systems. At present, power grid equipment management is completed through passive planned maintenance, ignoring the assessment and management of equipment health. With the geometric growth of power grid asset measurement data, equipment monitoring data, and various management system data, power grid companies need to use data-driven maintenance business plans and asset replacement throughout the life cycle of equipment to ensure safe, reliable and economical operation of the system.
目前对电网资产进行管理时，忽略对设备健康状况的评估与管理，管理效率低，自动化程度较差。Currently, when managing power grid assets, the assessment and management of equipment health are ignored, resulting in low management efficiency and poor automation.
发明内容Summary of the invention
本发明的主要目的在于克服现有技术的缺点与不足，提供基于案例驱动的电网资产全寿命管理方法及系统，通过建立三维案例管理数据库，在数据库中检索与新案例相似的案例，确定并返回相似度较高的相似案例集及其对应的解决方案，以及将相似度低于阈值的新案例添加至案例管理数据库中，实现对管理库的更新，管理效率高，自动化程度较好。The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and to provide a case-driven full-life cycle management method and system for power grid assets. By establishing a three-dimensional case management database, cases similar to new cases are retrieved from the database, a set of similar cases with high similarity and their corresponding solutions are determined and returned, and new cases with similarity below a threshold are added to the case management database, thereby realizing the update of the management library, with high management efficiency and a good degree of automation.
根据本发明的一个方面，本发明提供了一种基于案例驱动的电网资产全寿命管理方法，所述方法包括以下步骤：According to one aspect of the present invention, the present invention provides a case-driven full life cycle management method for power grid assets, the method comprising the following steps:
S1:提取电网资产管理历史案例数据，建立三维案例管理数据库，所述案例管理数据库包括案例空间和解决方案空间；S1: extract historical case data of power grid asset management and establish a three-dimensional case management database, wherein the case management database includes a case space and a solution space;
S2:将历史案例转换为相应的特性数据，并对历史案例转换后的特性数据进行规范化学习，对历史案例转换后的特性数据进行规范化学习包括：对非结构化数据进行标准转化及数据特征提取；S2: converting historical cases into corresponding characteristic data, and performing standardized learning on the characteristic data after the historical cases are converted. The standardized learning on the characteristic data after the historical cases are converted includes: performing standard transformation on unstructured data and extracting data features;
S3:进行新案例非结构化及非标准化数据在线识别及特征提取，提取新案例的特性，利用所提取的特性在所述案例管理数据库中检索相似案例，确定所述新案例与所述相似案例的相似度大于或等于第一阈值的相似案例集，返回所述相似案例集及其对应的解决方案；S3: Performing online identification and feature extraction of unstructured and non-standardized data of new cases, extracting characteristics of new cases, searching similar cases in the case management database using the extracted characteristics, determining a similar case set whose similarity between the new case and the similar case is greater than or equal to a first threshold, and returning the similar case set and its corresponding solution;
S4:确定所述新案例与所述相似案例的相似度是否小于或等于第二阈值，若是，将所述新案例添加至所述案例管理数据库中。S4: Determine whether the similarity between the new case and the similar case is less than or equal to a second threshold; if so, add the new case to the case management database.
优选的，本发明还包括以下步骤：Preferably, the present invention further comprises the following steps:
S5:通过潮流计算验证新案例的适用性。S5: Verify the applicability of the new case through power flow calculation.
优选地，采用权重最近邻法(WNN)进行案例检索，通过Euclidean距离公式来计算：Preferably, the weighted nearest neighbor method (WNN) is used for case retrieval, which is calculated by the Euclidean distance formula:
其中u表示案例特性，而v则是该案例特性总数；Where u represents the case characteristics, and v is the total number of characteristics of the case;
利用如下公式将Euclidean距离最短的案例聚类为一类，其中ck表示案例：The cases with the shortest Euclidean distance are clustered into one category using the following formula, where ck represents the case:
优选地，设置目标函数，计算新案例与已检索出的案例之间的误差，确定出相似案例，其中，目标函数为：Preferably, an objective function is set to calculate the error between the new case and the retrieved case to determine the similar case, wherein the objective function is:
利用如下公式计算误差：The error is calculated using the following formula:
其中，ωi为相应特征对应的权重值，而Rj则为检索出案例的特性值，εj则为对应的误差。Among them, ωi is the weight value corresponding to the corresponding feature, Rj is the characteristic value of the retrieved case, and εj is the corresponding error.
优选的，对非结构化数据进行标准转化，通过java.io流对数据读取，应用Jacob(Java-COM Bridge)及Java Excel API工具写入XML格式文档中。Preferably, the unstructured data is converted to a standard format, read through a java.io stream, and written into an XML format document using Jacob (Java-COM Bridge) and Java Excel API tools.
优选的，S3中进行新案例非结构化及非标准化数据在线识别及特征提取具体包括：应用XSD(XML schema definition)建立关系模型，建立结构映射和语义映射，为异构数据在目标关系数据库中实现完整有效的转换。利用自然语言解析的技术，实现对资产管理案例进行快速分析识别，为案例检索奠定基础。Preferably, the online identification and feature extraction of unstructured and non-standardized data of new cases in S3 specifically include: applying XSD (XML schema definition) to establish a relational model, establish structural mapping and semantic mapping, and achieve complete and effective conversion of heterogeneous data in the target relational database. Using natural language parsing technology, rapid analysis and identification of asset management cases can be achieved, laying the foundation for case retrieval.
优选的，所述三维是指时间维、逻辑维和知识维；在时间维，对设备从规划、采购、运检到退役进行全寿命状态跟踪，并用唯一编码进行身份及事件标注；在逻辑维，记录设备类型、专业用途、电压等级、正常运行方式以及家族关联设备信息；在知识维，记录设备相关检修、维护、案例更新信息；所述案例空间包括新案例和已解决案例；所述解决方案空间包括历史解决方案和自学习解决方案。Preferably, the three dimensions refer to the time dimension, the logic dimension and the knowledge dimension; in the time dimension, the equipment is tracked throughout its life cycle from planning, procurement, operation and inspection to retirement, and its identity and events are marked with unique codes; in the logic dimension, the equipment type, professional use, voltage level, normal operation mode and family-related equipment information are recorded; in the knowledge dimension, equipment-related inspection, maintenance and case update information are recorded; the case space includes new cases and resolved cases; and the solution space includes historical solutions and self-learning solutions.
优选地，对从案例库中检索出的案例按相关性进行筛选和排序，并对相关性大于阈值的案例进行安全性验证，以计算在检修条件下，系统的安全性和可靠性，以及预计由于检修所造成的断电时间及相应的经济损失。Preferably, the cases retrieved from the case library are screened and sorted according to relevance, and the cases with relevance greater than a threshold are safety verified to calculate the safety and reliability of the system under maintenance conditions, as well as the expected power outage time and corresponding economic losses caused by the maintenance.
根据本发明的另一方面，本发明还提供了一种基于案例驱动的电网资产全寿命管理系统，所述系统包括：According to another aspect of the present invention, the present invention further provides a case-driven power grid asset lifecycle management system, the system comprising:
建立模块，用于提取电网资产管理历史案例数据，建立三维案例管理数据库，所述案例管理数据库包括案例空间和解决方案空间；Establishing a module for extracting historical case data of power grid asset management and establishing a three-dimensional case management database, wherein the case management database includes a case space and a solution space;
转换模块，将历史案例转换为相应的特性数据，并对历史案例转换后的特性数据进行规范化学习，对历史案例转换后的特性数据进行规范化学习包括：对非结构化数据进行标准转化及数据特征提取；The conversion module converts historical cases into corresponding characteristic data and performs standardized learning on the characteristic data after the historical cases are converted. The standardized learning on the characteristic data after the historical cases are converted includes: standard transformation of unstructured data and data feature extraction;
检索模块，用于进行新案例非结构化及非标准化数据在线识别及特征提取，提取新案例的特性，利用所提取的特性在所述案例管理数据库中检索相似案例，确定所述新案例与所述相似案例的相似度大于或等于第一阈值的相似案例集，返回所述相似案例集及其对应的解决方案；A retrieval module, used for online identification and feature extraction of unstructured and non-standardized data of new cases, extracting characteristics of new cases, searching similar cases in the case management database using the extracted characteristics, determining a set of similar cases whose similarity between the new case and the similar cases is greater than or equal to a first threshold, and returning the set of similar cases and their corresponding solutions;
确定模块，用于确定所述新案例与所述相似案例的相似度是否小于或等于第二阈值，若是，将所述新案例添加至所述案例管理数据库中。The determination module is used to determine whether the similarity between the new case and the similar case is less than or equal to a second threshold, and if so, add the new case to the case management database.
优选地，对于新案例请求中非结构化数据和非规范化数据进行快速分析识别；采用运行知识库信息，对大量设备运行数据应用聚类算法和语义识别算法，对调度特征信息进行深度挖掘及统计分析：Preferably, unstructured data and non-standardized data in new case requests are quickly analyzed and identified; clustering algorithms and semantic recognition algorithms are applied to a large amount of equipment operation data using operation knowledge base information, and scheduling feature information is deeply mined and statistically analyzed:
优选地，采用权重最近邻法(WNN)进行案例检索，通过Euclidean距离公式来计算：Preferably, the weighted nearest neighbor method (WNN) is used for case retrieval, which is calculated by the Euclidean distance formula:
其中u表示案例特性，而v则是该案例特性总数；Where u represents the case characteristics, and v is the total number of characteristics of the case;
利用如下公式将Euclidean距离最短的案例聚类为一类，其中ck表示案例：The cases with the shortest Euclidean distance are clustered into one category using the following formula, where ck represents the case:
优选地，设置目标函数，计算新案例与已检索出的案例之间的误差，确定出相似案例，其中，目标函数为：Preferably, an objective function is set to calculate the error between the new case and the retrieved case to determine the similar case, wherein the objective function is:
利用如下公式计算误差：The error is calculated using the following formula:
其中，ωi为相应特征对应的权重值，而Rj则为检索出案例的特性值，εj则为对应的误差。Among them, ωi is the weight value corresponding to the corresponding feature, Rj is the characteristic value of the retrieved case, and εj is the corresponding error.
优选地，所述三维是指时间维、逻辑维和知识维；在时间维，对设备从规划、采购、运检到退役进行全寿命状态跟踪，并用唯一编码进行身份及事件标注；在逻辑维，记录设备类型、专业用途、电压等级、正常运行方式以及家族关联设备信息；在知识维，记录设备相关检修、维护、案例更新信息；所述案例空间包括新案例和已解决案例；所述解决方案空间包括历史解决方案和自学习解决方案。Preferably, the three dimensions refer to the time dimension, the logic dimension and the knowledge dimension; in the time dimension, the equipment is tracked throughout its life cycle from planning, procurement, operation and inspection to retirement, and its identity and events are marked with unique codes; in the logic dimension, the equipment type, professional use, voltage level, normal operation mode and family-related equipment information are recorded; in the knowledge dimension, equipment-related inspection, maintenance and case update information are recorded; the case space includes new cases and resolved cases; and the solution space includes historical solutions and self-learning solutions.
优选地，对新案例进行特征提取，而后对从案例库中检索出的案例按相关性进行筛选和排序，并对相关性大于阈值的案例进行安全性验证，以计算在检修条件下，系统的安全性和可靠性，以及预计由于检修所造成的断电时间及相应的经济损失。Preferably, feature extraction is performed on new cases, and then the cases retrieved from the case library are screened and sorted according to relevance, and safety verification is performed on cases with relevance greater than a threshold, so as to calculate the safety and reliability of the system under maintenance conditions, as well as the expected power outage time and corresponding economic losses caused by the maintenance.
有益效果：本发明通过建立三维案例管理数据库，在数据库中检索与新案例相似的案例，确定并返回相似度较高的相似案例集及其对应的解决方案，以及将相似度低于阈值的新案例添加至案例管理数据库中，实现对管理库的更新，管理效率高，自动化程度较好。Beneficial effects: The present invention establishes a three-dimensional case management database, retrieves cases similar to new cases in the database, determines and returns a set of similar cases with high similarity and their corresponding solutions, and adds new cases with similarity below a threshold to the case management database, thereby updating the management library with high management efficiency and a good degree of automation.
通过参照以下附图及对本发明的具体实施方式的详细描述，本发明的特征及优点将会变得清楚。The features and advantages of the present invention will become clear through reference to the following drawings and detailed description of specific embodiments of the present invention.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1是电网资产全寿命管理方法流程图；FIG1 is a flow chart of a method for managing the entire life cycle of power grid assets;
图2是管理数据库的三维示意图；FIG2 is a three-dimensional schematic diagram of a management database;
图3是管理数据库的结构示意图；FIG3 is a schematic diagram of the structure of a management database;
图4是案例数据特征提取示意图；FIG4 is a schematic diagram of case data feature extraction;
图5是资产管理需求语义识别示意图；Figure 5 is a schematic diagram of semantic recognition of asset management requirements;
图6是案例检索流程图；Fig. 6 is a case search flow chart;
图7是电网资产全寿命管理系统示意图。FIG7 is a schematic diagram of the power grid asset life cycle management system.
具体实施方式DETAILED DESCRIPTION
下面结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，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。The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
实施例1Example 1
图1是电网资产全寿命管理方法流程图。如图1所示，本发明提供一种基于案例驱动的电网资产全寿命管理方法，所述方法包括以下步骤：FIG1 is a flow chart of a method for managing the life cycle of power grid assets. As shown in FIG1 , the present invention provides a method for managing the life cycle of power grid assets based on case driving, and the method comprises the following steps:
S1:建立三维案例管理数据库，所述案例管理数据库包括案例空间和解决方案空间。S1: Establish a three-dimensional case management database, which includes a case space and a solution space.
如图2所示，所述三维是指时间维、逻辑维和知识维；在时间维，对设备从规划、采购、运检到退役进行全寿命状态跟踪，并用唯一编码进行身份及事件标注；在逻辑维，记录设备类型、专业用途、电压等级、正常运行方式以及家族关联设备信息；在知识维，记录设备相关检修、维护、案例更新信息。该系统从设备的全寿命周期出发，应用云端同步技术，根据各个维度中子维度数据在三维空间中形成该设备生命周期内的相关事件逻辑关系及事件库，对设备进行全时空维度上的跟踪画像，为设备检修维护提供辅助决策。As shown in Figure 2, the three dimensions refer to the time dimension, the logic dimension, and the knowledge dimension. In the time dimension, the equipment is tracked throughout its life cycle from planning, procurement, operation and inspection to retirement, and its identity and events are marked with a unique code. In the logic dimension, the equipment type, professional use, voltage level, normal operation mode, and family-related equipment information are recorded. In the knowledge dimension, the equipment-related overhaul, maintenance, and case update information are recorded. Starting from the full life cycle of the equipment, the system uses cloud synchronization technology to form the logical relationship and event library of related events in the life cycle of the equipment in three-dimensional space based on the sub-dimensional data of each dimension, and tracks and portrays the equipment in all time and space dimensions, providing auxiliary decision-making for equipment overhaul and maintenance.
其中，电网资产全寿命管理逻辑维数据信息表如表1所示：Among them, the data information table of the logic dimension of the full life cycle management of power grid assets is shown in Table 1:
表1：电网资产全寿命管理逻辑维数据信息表Table 1: Logical dimension data information table of power grid asset life cycle management
如图3所示，所述案例空间包括新案例和已解决案例；所述解决方案空间包括专家库解决方案和自学习解决方案。As shown in FIG3 , the case space includes new cases and solved cases; the solution space includes expert library solutions and self-learning solutions.
S2:将历史案例转换为相应的特性数据，并对历史案例转换后的特性数据进行规范化学习，对历史案例转换后的特性数据进行规范化学习包括：对非结构化数据进行标准转化及数据特征提取(图4)；S2: converting historical cases into corresponding characteristic data, and performing standardized learning on the characteristic data after the historical cases are converted. The standardized learning on the characteristic data after the historical cases are converted includes: performing standard transformation on unstructured data and extracting data features (Figure 4);
S3:进行新案例非结构化及非标准化数据在线识别及特征提取，提取新案例的特性，利用所述特性在所述案例管理数据库中检索相似案例，确定所述新案例与所述相似案例的相似度大于或等于第一阈值的相似案例集，返回所述相似案例集及其对应的解决方案。S3: Perform online identification and feature extraction of unstructured and non-standardized data of new cases, extract characteristics of new cases, use the characteristics to retrieve similar cases in the case management database, determine a set of similar cases whose similarity between the new case and the similar cases is greater than or equal to a first threshold, and return the set of similar cases and their corresponding solutions.
案例可以表示为C＝(P,S,T)，其中P表示在搜索空间上对问题的描述，即The case can be expressed as C = (P, S, T), where P represents the description of the problem in the search space, that is,
P＝{Agent:[u1,u2,……,uv]} (1)P＝{Agent:[u1 ,u2 ,...,uv ]} (1)
对于问题的描述P，可以表示为每个代理agent在搜索空间上的主体特性ui的集合，v则是特性总数。S表示解决方案及相应步骤的集合，T则是相应的信任度值。The description of the problem P can be expressed as a set of principal features ui of each agent in the search space, and v is the total number of features. S represents the set of solutions and corresponding steps, and T is the corresponding trust value.
图6是案例检索流程图。权重最近邻法(Weighted Nearest Neighbour,WNN)是简单直观的检索方法。当数据源为数据或可转换为数据的文件时，应采用WNN对云端案例库进行检索，通常应用Euclidean距离公式来计算：Figure 6 is a case retrieval flow chart. Weighted Nearest Neighbor (WNN) is a simple and intuitive retrieval method. When the data source is data or a file that can be converted into data, WNN should be used to search the cloud case library, usually using the Euclidean distance formula to calculate:
其中u表示案例特性，而v则是该案例特性总数。利用如公式(3)所示的聚类方法，将 Euclidean距离最短的案例聚类为一类：Where u represents the case characteristics, and v is the total number of case characteristics. Using the clustering method shown in formula (3), the cases with the shortest Euclidean distance are clustered into one category:
其中ck为案例库中某一案例。Where ck is a case in the case library.
为更好的观测新的案例与已检索出案例之间的误差，将二者进行比较，并利用公式(4) 计算其误差。In order to better observe the error between the new case and the retrieved case, the two are compared and their error is calculated using formula (4).
其中，ωi为相应特征对应的权重值，而Rj则为检索出案例的特性值，εj则为对应的误差。为了减少误差εj，设定目标函数为：Among them, ωi is the weight value corresponding to the corresponding feature, Rj is the characteristic value of the retrieved case, and εj is the corresponding error. In order to reduce the error εj , the objective function is set as:
采用支持向量机(Support Vector Machine,SVM)来解决公式(5)的优化问题。SVM采用高斯径向基核函数(Radial Basis Function，RBF)作为核函数，表示u到核函数中心uc之间的欧式距离。记作：Support Vector Machine (SVM) is used to solve the optimization problem of formula (5). SVM uses Gaussian Radial Basis Function (RBF) as the kernel function, which represents the Euclidean distance between u and the kernel function center uc . It is denoted as:
其中σ为该函数的宽度参数，以控制函数的径向作用范围。Where σ is the width parameter of the function, which controls the radial range of the function.
在一个实施例中，在进行检索时，首先进行设备信息识别，如系统信息为“变电站ABC 220kV-1母线：x出口线路维护”。系统首先检索变电站ABC 220kV-1母线下所有出口设备类型、检修情况、历史事件，形成自身及相关设备特征点提取以及历史事件特征提取，具体检索结果如表2所示：In one embodiment, when searching, the device information is first identified. For example, if the system information is "Substation ABC 220kV-1 bus: x export line maintenance", the system first searches for all export equipment types, maintenance conditions, and historical events under the substation ABC 220kV-1 bus, and extracts feature points of itself and related equipment as well as historical event features. The specific search results are shown in Table 2:
表2：基于云的案例检索结果表Table 2: Cloud-based case search results
在实际系统中，特征点提取过程中，由于系统输入内容的人为因素造成的不规范信息，给案例检索造成很大影响，系统通过非结构化数据语义识别加以甄别(图5)。如：“变电站ABC”可能检索出“电站ABC”，同时系统中可能存在类似如：“ABC电厂”及“ABC馈线”等类似信息，同样给检索造成混淆，通常定义检索关键词顺序以完成快速精准检索。In the actual system, during the feature point extraction process, the non-standard information caused by human factors in the system input content has a great impact on case retrieval. The system identifies it through unstructured data semantic recognition (Figure 5). For example, "ABC substation" may retrieve "ABC power station". At the same time, there may be similar information such as "ABC power plant" and "ABC feeder" in the system, which also confuses the search. Usually, the search keyword order is defined to complete fast and accurate search.
优选地，对从案例库中检索出的案例按相关性进行筛选和排序，并对相关性大于阈值的案例进行安全性验证，以计算在检修条件下，系统的安全性和可靠性，以及预计由于检修所造成的断电时间及相应的经济损失。Preferably, the cases retrieved from the case library are screened and sorted according to relevance, and the cases with relevance greater than a threshold are safety verified to calculate the safety and reliability of the system under maintenance conditions, as well as the expected power outage time and corresponding economic losses caused by the maintenance.
如表2所示，案例库检索案例按相关性进行筛选和排序，并将其中相关性最强的5条案例进行安全性验证，以计算在检修条件下，系统的安全性和可靠性，以及预计由于检修所造成的断电时间及相应的经济损失，并按照公式(5)计算得出的评分进行重新排序。从而可重新得到检修任务中的整体设备信息，包括：变电站ABC 220kV-1母线：x/y/z出口；变电站ABC 220kV-2母线:s出口；变电站ABC 220kV-3母线。进行厂站、设备相关信息的提取，得到：变电站ABC 220kV-1母线；相关设备：出口线路、变压器；历史检修信息汇总：出口线路、变压器更换<时间label>；调控单位：市调；设备归属：高压输电管理；检修方案：对变电站ABC 220kV-1母线停电，由变电站ABC 220kV-2母线转供，停电时间2小时。As shown in Table 2, the cases retrieved from the case library are screened and sorted by relevance, and the five most relevant cases are safety verified to calculate the safety and reliability of the system under maintenance conditions, as well as the expected power outage time and corresponding economic losses caused by maintenance, and are re-sorted according to the scores calculated by formula (5). In this way, the overall equipment information in the maintenance task can be obtained again, including: substation ABC 220kV-1 bus: x/y/z outlet; substation ABC 220kV-2 bus: s outlet; substation ABC 220kV-3 bus. The plant and equipment related information is extracted to obtain: substation ABC 220kV-1 bus; related equipment: export line, transformer; historical maintenance information summary: export line, transformer replacement <time label>; control unit: municipal control; equipment ownership: high-voltage transmission management; maintenance plan: power outage on substation ABC 220kV-1 bus, and transfer power from substation ABC 220kV-2 bus, with a power outage time of 2 hours.
系统检索出的案例将由调控人员进行判断并予以安排实施，实施成功后将以新案例的形式重新写入案例库，以对设备检修历史事件进行记录的同时也对案例库进行云端更新。The cases retrieved by the system will be judged and arranged for implementation by the control personnel. After successful implementation, they will be rewritten into the case library in the form of new cases to record the historical events of equipment maintenance and update the case library in the cloud.
S4:确定所述新案例与所述相似案例的相似度是否小于或等于第二阈值，若是，将所述新案例添加至所述案例管理数据库中。S4: Determine whether the similarity between the new case and the similar case is less than or equal to a second threshold; if so, add the new case to the case management database.
对于某些案例，系统会检索出多种相似度接近的解决方案。为此，从运行维护实际出发，为每种方法设置权重因子，特别是对于与重要负荷、资产相关的解决方案中，曾在资产管理历史中成功解决相似问题的方案，设置优先因子，相应的解决方案将作为首选项提供给调度运维管理中心。For some cases, the system will retrieve multiple solutions with similarities. To this end, based on the actual operation and maintenance, a weight factor is set for each method, especially for solutions related to important loads and assets, and priority factors are set for solutions that have successfully solved similar problems in the history of asset management. The corresponding solutions will be provided to the dispatching and operation management center as the first choice.
经过检索及自学习的新案例是否应该更新入云端案例库需要根据新案例与案例库已有案例的相似度进行判断，本实施例采用相似度90％作为判别标准，即：Whether the new cases after retrieval and self-learning should be updated into the cloud case library needs to be judged based on the similarity between the new cases and the existing cases in the case library. This embodiment uses 90% similarity as the judgment standard, that is:
similarity(NewCase,SimilarCases)≤90％ (7)similarity(NewCase,SimilarCases)≤90% (7)
如果(7)式成立，则将新案例添加至案例库中。If (7) holds true, the new case is added to the case library.
S5:通过潮流计算验证新案例的适用性。S5: Verify the applicability of the new case through power flow calculation.
本实施例通过建立三维案例管理数据库，在数据库中检索与新案例相似的案例，确定并返回相似度较高的相似案例集及其对应的解决方案，以及将相似度低于阈值的新案例添加至案例管理数据库中，实现对管理库的更新，管理效率高，自动化程度较好。This embodiment establishes a three-dimensional case management database, searches for cases similar to new cases in the database, determines and returns a set of similar cases with high similarity and their corresponding solutions, and adds new cases with similarity below a threshold to the case management database, thereby updating the management library with high management efficiency and a good degree of automation.
实施例2Example 2
图7是电网资产全寿命管理系统示意图。如图7所示，本发明还提供了一种基于案例驱动的电网资产全寿命管理系统，所述系统包括：FIG7 is a schematic diagram of a power grid asset lifecycle management system. As shown in FIG7 , the present invention further provides a case-driven power grid asset lifecycle management system, the system comprising:
建立模块，用于提取电网资产管理历史案例数据，建立三维案例管理数据库，所述案例管理数据库包括案例空间和解决方案空间；Establishing a module for extracting historical case data of power grid asset management and establishing a three-dimensional case management database, wherein the case management database includes a case space and a solution space;
转换模块，用于将历史案例转换为相应的特性数据，并对历史案例数据进行规范化学习，对历史案例数据进行规范化学习包括非结构化数据转化及数据特征提取；A conversion module is used to convert historical cases into corresponding characteristic data and perform normalized learning on the historical case data. Normalized learning on the historical case data includes unstructured data conversion and data feature extraction;
检索模块，用于进行新案例非结构化及非标准化数据在线识别及特征提取，提取新案例的特性，利用所提取的特性在所述案例管理数据库中检索相似案例，确定所述新案例与所述相似案例的相似度大于或等于第一阈值的相似案例集，返回所述相似案例集及其对应的解决方案；A retrieval module, used for online identification and feature extraction of unstructured and non-standardized data of new cases, extracting characteristics of new cases, searching similar cases in the case management database using the extracted characteristics, determining a set of similar cases whose similarity between the new case and the similar cases is greater than or equal to a first threshold, and returning the set of similar cases and their corresponding solutions;
确定模块，用于确定所述新案例与所述相似案例的相似度是否小于或等于第二阈值，若是，将所述新案例添加至所述案例管理数据库中。The determination module is used to determine whether the similarity between the new case and the similar case is less than or equal to a second threshold, and if so, add the new case to the case management database.
优选地，采用权重最近邻法(WNN)进行案例检索，通过Euclidean距离公式来计算：Preferably, the weighted nearest neighbor method (WNN) is used for case retrieval, which is calculated by the Euclidean distance formula:
其中u表示案例特性，而v则是该案例特性总数；Where u represents the case characteristics, and v is the total number of characteristics of the case;
利用如下公式将Euclidean距离最短的案例聚类为一类，其中ck表示案例：The cases with the shortest Euclidean distance are clustered into one category using the following formula, where ck represents the case:
优选地，设置目标函数，计算新案例与已检索出的案例之间的误差，确定出相似案例，其中，目标函数为：Preferably, an objective function is set to calculate the error between the new case and the retrieved case to determine the similar case, wherein the objective function is:
利用如下公式计算误差：The error is calculated using the following formula:
其中，ωi为相应特征对应的权重值，而Rj则为检索出案例的特性值，εj则为对应的误差。Among them, ωi is the weight value corresponding to the corresponding feature, Rj is the characteristic value of the retrieved case, and εj is the corresponding error.
优选地，所述三维是指时间维、逻辑维和知识维；在时间维，将设备从规划、采购、运检到退役进行状态标记，并用唯一编码进行身份标注；在逻辑维，记录设备类型、专业用途、电压等级、正常运行方式以及关联设备信息；在知识维，记录设备相关检修、维护、更新信息；所述案例空间包括新案例和已解决案例；所述解决方案空间包括专家库解决方案和自学习解决方案。Preferably, the three dimensions refer to the time dimension, the logic dimension and the knowledge dimension; in the time dimension, the status of the equipment is marked from planning, procurement, operation and inspection to retirement, and the identity is marked with a unique code; in the logic dimension, the equipment type, professional use, voltage level, normal operation mode and related equipment information are recorded; in the knowledge dimension, the equipment-related inspection, maintenance and update information is recorded; the case space includes new cases and solved cases; the solution space includes expert library solutions and self-learning solutions.
优选地，对从案例库中检索出的案例按相关性进行筛选和排序，并对相关性大于阈值的案例进行安全性验证，以计算在检修条件下，系统的安全性和可靠性，以及预计由于检修所造成的断电时间及相应的经济损失。Preferably, the cases retrieved from the case library are screened and sorted according to relevance, and the cases with relevance greater than a threshold are safety verified to calculate the safety and reliability of the system under maintenance conditions, as well as the expected power outage time and corresponding economic losses caused by the maintenance.
本实施例2中各个模块所执行的方法步骤的具体实施过程与实施例1中的各个步骤的实施过程相同，在此不再赘述。The specific implementation process of the method steps executed by each module in this embodiment 2 is the same as the implementation process of each step in embodiment 1, and will not be repeated here.
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 on 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 implemented 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 implementing 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 |
|---|---|---|---|
| CN202110502223.8ACN113191735B (en) | 2021-05-08 | 2021-05-08 | Case-driven-based power grid asset life-span management method and system |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110502223.8ACN113191735B (en) | 2021-05-08 | 2021-05-08 | Case-driven-based power grid asset life-span management method and system |
| Publication Number | Publication Date |
|---|---|
| CN113191735A CN113191735A (en) | 2021-07-30 |
| CN113191735Btrue CN113191735B (en) | 2024-08-13 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110502223.8AActiveCN113191735B (en) | 2021-05-08 | 2021-05-08 | Case-driven-based power grid asset life-span management method and system |
| Country | Link |
|---|---|
| CN (1) | CN113191735B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110059052A (en)* | 2019-02-28 | 2019-07-26 | 石化盈科信息技术有限责任公司 | A kind of refinery scheduling case management method and computer readable storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104851025A (en)* | 2015-05-09 | 2015-08-19 | 湘南学院 | Case-reasoning-based personalized recommendation method for E-commerce website commodity |
| CN111221920B (en)* | 2019-11-29 | 2022-11-01 | 国网电力科学研究院武汉南瑞有限责任公司 | Case base construction method and device for power transformation equipment operation and maintenance device and computer storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110059052A (en)* | 2019-02-28 | 2019-07-26 | 石化盈科信息技术有限责任公司 | A kind of refinery scheduling case management method and computer readable storage medium |
| Publication number | Publication date |
|---|---|
| CN113191735A (en) | 2021-07-30 |
| Publication | Publication Date | Title |
|---|---|---|
| CN110751317A (en) | Power load prediction system and prediction method | |
| CN118410405A (en) | Intelligent identification system for hierarchical relationship of data assets | |
| CN115269578A (en) | Data index-based comprehensive energy enterprise data management method and system | |
| CN118227599A (en) | Data standard treatment method, system, equipment and medium based on automatic flow | |
| CN116306923A (en) | Evaluation weight calculation method based on knowledge graph | |
| CN118193747A (en) | Electric power knowledge graph construction method, system, equipment and medium | |
| CN111061679A (en) | Method and system for rapid configuration of technological innovation policy based on rete and drools rules | |
| CN118070784A (en) | Method, device, equipment and storage medium for constructing entity dictionary in vertical industry field | |
| CN115344717B (en) | Method and device for constructing knowledge map for multi-type energy supply and utilization system regulation and operation | |
| CN114491168B (en) | Regulating cloud sample data sharing method, system, computer equipment and storage medium | |
| CN113191735B (en) | Case-driven-based power grid asset life-span management method and system | |
| CN119669403A (en) | Power grid marketing data annotation method based on perplexity-driven large language model | |
| Bella et al. | Semi-supervised approach for recovering traceability links in complex systems | |
| CN118627752A (en) | A data processing system based on the big data platform of the entire tin industry chain | |
| CN118410866A (en) | A modeling method for regional main network knowledge graph model | |
| Li et al. | Automatic classification algorithm for multisearch data association rules in wireless networks | |
| CN116882820A (en) | Situation analysis method and device for electric power marketing and computer equipment | |
| CN115688729A (en) | A cost data integration management system and method for power transmission and transformation projects | |
| CN115759562A (en) | Logic relation and construction period calculation algorithm of preventive maintenance work order for overhaul of nuclear power plant | |
| CN111428756B (en) | Real-time method and device for planning data fusion based on time series information entropy | |
| CN119862286B (en) | Knowledge graph management method and system for smelting industry | |
| CN119377412B (en) | Service work order classification method, electronic device, storage medium, and program product | |
| Li et al. | High-quality sample generation for power system transient stability assessment based on data-driven methods | |
| CN120297904B (en) | Comprehensive consulting services and project management platform for the transformation of scientific and technological achievements | |
| CN117874240B (en) | Audit text classification method, system and equipment based on knowledge graph |
| 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 |