技术领域Technical field
本发明涉及电网数据处理技术领域,具体地说是一种电网薄弱环节动态监测预警决策系统及方法。The invention relates to the technical field of power grid data processing, specifically a dynamic monitoring and early warning decision-making system and method for weak links in the power grid.
背景技术Background technique
“实时排查、精准诊断、高效治理”电网薄弱环节是电网高质量发展的基础。电网发展诊断分析所需的电网、用户基础数据规模大、涉及领域广、参与部门多、更新变化快，传统依靠人工方式核查海量电网数据，从数据中逐条筛查电网薄弱环节，存在薄弱环节感知难、定位难、治理难等问题，难以适应电网精益化管理、精准化投资的要求，需要有效的信息化工具集成、处理、共享数据并支撑规划项目安排等业务决策。"Real-time investigation, accurate diagnosis, and efficient management" of weak links in the power grid are the foundation for the high-quality development of the power grid. The basic data of power grid and users required for diagnosis and analysis of power grid development are large in scale, involve a wide range of fields, involve many departments, and update and change rapidly. Traditionally, manual verification is relied on to check massive power grid data, and the weak links of the power grid are screened one by one from the data to detect the existence of weak links. Problems such as difficulty, positioning, and governance make it difficult to adapt to the requirements of lean management and precise investment in the power grid. Effective information tools are needed to integrate, process, and share data and support business decisions such as planning project arrangements.
电网薄弱环节动态监测预警主要是单项内容，如可靠性、安全性和供电质量等方面，不能完全覆盖电网规划所需信息，难以全面指导实际的电网建设。在电网建设发展方面，已有一些电网薄弱环节指标体系可供参考，但其主要是依据当前能够直接获得的电网数据而确定，受数据传递壁垒等因素制约，导致预警可信度低。Dynamic monitoring and early warning of weak links in the power grid mainly focus on single items, such as reliability, security, and power supply quality. It cannot fully cover the information required for power grid planning, and it is difficult to comprehensively guide actual power grid construction. In terms of power grid construction and development, there are already some power grid weak link indicator systems for reference, but they are mainly determined based on the power grid data that can be directly obtained at present. They are restricted by data transmission barriers and other factors, resulting in low credibility of early warnings.
发明内容Contents of the invention
本发明实施例中提供了一种电网薄弱环节动态监测预警决策系统及方法，以解决现有技术中电网预警的参考依据单一、预警结果可信度低的问题。Embodiments of the present invention provide a dynamic monitoring and early warning decision-making system and method for weak links in the power grid to solve the problems in the prior art that the reference basis for power grid early warning is single and the credibility of the early warning results is low.
为了解决上述技术问题，本发明实施例公开了如下技术方案：In order to solve the above technical problems, embodiments of the present invention disclose the following technical solutions:
本发明第一方面提供了一种电网薄弱环节动态监测预警决策系统，所述系统包括：The first aspect of the present invention provides a dynamic monitoring and early warning decision-making system for weak links in the power grid. The system includes:
电网信息感知模块，通过ETL或服务调用的方式接入电力业务系统，获取业务系统数据，并通过网络爬虫获取用电量关联信息，形成电网规划综合数据库，并建立数据库内多源数据的关联关系，进行数据融合和电网设备建模；The power grid information sensing module is connected to the power business system through ETL or service invocation, obtains business system data, and obtains electricity consumption related information through web crawlers, forms a comprehensive database for power grid planning, and establishes correlation relationships between multi-source data in the database. , perform data fusion and power grid equipment modeling;
电网状态诊断模块，建立电网诊断专家知识库，从多个维度定义电网薄弱环节指标，基于所述电网规划综合数据库，通过多算法引擎，实时计算所述电网薄弱环节的指标值，推送问题清单；The power grid status diagnosis module establishes a power grid diagnosis expert knowledge base, defines power grid weak link indicators from multiple dimensions, and based on the power grid planning comprehensive database, uses a multi-algorithm engine to calculate the indicator values of the power grid weak links in real time and push a list of problems;
薄弱环节预警模块，基于所述电网规划综合数据库，建立参数档案和运行档案，结合影响因素，建立预警模型，采用人工智能或深度学习算法预测预警，形成预警清单；The weak link early warning module is based on the power grid planning comprehensive database, establishes parameter files and operation files, combines influencing factors, establishes an early warning model, uses artificial intelligence or deep learning algorithms to predict early warnings, and forms an early warning list;
智能辅助决策模块，对于所述问题清单和预警清单，提出解决方案，并向人工发送处理工单，收集人工处理进程及处理结果。The intelligent auxiliary decision-making module proposes solutions to the problem list and early warning list, sends processing work orders to manual processing, and collects the manual processing progress and processing results.
进一步地，所述电网信息感知模块包括：Further, the power grid information sensing module includes:
模型匹配计算单元，建立多源数据的特征属性在内的数据模型，通过模型内属性信息的遍历对比，计算模型间的匹配度；The model matching calculation unit establishes a data model including the characteristic attributes of multi-source data, and calculates the matching degree between models through traversal and comparison of attribute information in the model;
多源数据融合单元，根据所述匹配度的计算情况自动关联，实现多源数据自动集成融合。The multi-source data fusion unit automatically correlates according to the calculation of the matching degree to realize automatic integration and fusion of multi-source data.
本发明第二方面提供了一种电网薄弱环节动态监测预警决策方法，所述方法包括以下步骤：A second aspect of the present invention provides a dynamic monitoring and early warning decision-making method for weak links in the power grid. The method includes the following steps:
通过ETL或服务调用的方式接入电力业务系统，获取业务系统数据，并通过网络爬虫获取用电量关联信息，形成电网规划综合数据库，并建立数据库内多源数据的关联关系，进行数据融合和电网设备建模；Access the power business system through ETL or service invocation, obtain business system data, and obtain electricity consumption related information through web crawlers to form a comprehensive database for power grid planning, and establish correlation relationships between multi-source data in the database to perform data fusion and Grid equipment modeling;
建立电网诊断专家知识库，从多个维度定义电网薄弱环节指标，基于所述电网规划综合数据库，通过多算法引擎，实时计算所述电网薄弱环节的指标值，推送问题清单；Establish a knowledge base for power grid diagnosis experts to define indicators of power grid weak links from multiple dimensions. Based on the power grid planning comprehensive database and through a multi-algorithm engine, calculate the indicator values of the power grid weak links in real time and push a list of issues;
基于所述电网规划综合数据库，建立参数档案和运行档案，结合影响因素，建立预警模型，采用人工智能或深度学习算法预测预警，形成预警清单；Based on the power grid planning comprehensive database, establish parameter files and operation files, combine influencing factors, establish an early warning model, use artificial intelligence or deep learning algorithms to predict early warnings, and form an early warning list;
对于所述问题清单和预警清单，提出解决方案，并向人工发送处理工单，收集人工处理进程及处理结果。Propose solutions to the problem list and early warning list, send manual processing work orders, and collect manual processing progress and processing results.
进一步地，所述电力业务系统包括PMS系统、GIS系统、EMS系统、供电服务指挥系统和用电信息采集系统；所述用电量关联信息包括经济信息、人口信息、能源信息和政府规划信息。Further, the power business system includes a PMS system, a GIS system, an EMS system, a power supply service command system and a power consumption information collection system; the power consumption related information includes economic information, population information, energy information and government planning information.
进一步地，所述建立数据库内多源数据的关联关系，进行数据融合和电网设备建模的具体过程为：Further, the specific process of establishing the association of multi-source data in the database, performing data fusion and modeling of power grid equipment is as follows:
建立包含电力业务系统的数据逻辑关系、拓扑结构、空间信息和特征属性的模型；Establish a model containing the data logical relationships, topological structure, spatial information and characteristic attributes of the power business system;
通过模型内的属性信息进行遍历对比，计算模型间匹配度；Traverse and compare the attribute information in the model to calculate the matching degree between models;
根据匹配度情况进行模型关联，实现数据的集成融合；Carry out model association according to the matching degree to achieve data integration and fusion;
对于未能融合的数据，按数据所属电压等级和/或设备所在区域进行人工处理融合。For data that cannot be fused, manual processing and fusion will be performed based on the voltage level to which the data belongs and/or the area where the equipment is located.
进一步地，所述模型间匹配度的计算依次通过根节点对应、逻辑模型遍历、有向图遍历、特征属性识别、空间坐标转换和图像识别的方式进行。Further, the calculation of the matching degree between the models is performed sequentially through root node correspondence, logical model traversal, directed graph traversal, feature attribute identification, spatial coordinate conversion and image recognition.
进一步地，所述算法引擎包括拓扑识别、N-1计算、负载分析、容载比计算、供电范围识别、运行方式识别。Further, the algorithm engine includes topology identification, N-1 calculation, load analysis, capacity-to-load ratio calculation, power supply range identification, and operating mode identification.
进一步地，所述基于所述电网规划综合数据库，通过多算法引擎，实时计算所述电网薄弱环节的指标值，推送问题清单的具体过程为：Further, based on the power grid planning comprehensive database, through a multi-algorithm engine, the indicator values of the power grid weak links are calculated in real time, and the specific process of pushing the problem list is:
依据电网薄弱环节指标判定标准，定义指标计算公式，设置指标阈值；Based on the indicator criteria for determining weak links in the power grid, define the indicator calculation formula and set the indicator threshold;
利用分布式并行计算和数据探针技术，基于电网规划综合数据库内的数据，通过算法引擎，依据维度诊断体系及计算规则，实时计算指标得分，动态监测电网设备运行状态，推送包括设备问题、成因问题、指标得分、空间位置和问题紧迫程度在内的问题清单。Utilize distributed parallel computing and data probe technology, based on the data in the power grid planning comprehensive database, through the algorithm engine, according to the dimensional diagnosis system and calculation rules, real-time calculation of indicator scores, dynamic monitoring of the operating status of power grid equipment, and push notifications including equipment problems and causes List of issues including issue, metric score, spatial location, and issue urgency.
进一步地，所述采用深度学习算法预测预警的具体过程为：Further, the specific process of using deep learning algorithm to predict and warn is as follows:
设置要预测模型或参数的经验值；Set the empirical value of the model or parameter to be predicted;
针对设定历史时间段内产生的实际值，逐一预测，将预测结果与实际发生结果进行拟合度对比，对每个参数按照设定精度进行调整；Forecast the actual values generated within the set historical time period one by one, compare the fit between the predicted results and the actual results, and adjust each parameter according to the set accuracy;
利用调整结果进行问题预警。Use the adjustment results to provide early warning of problems.
发明内容中提供的效果仅仅是实施例的效果，而不是发明所有的全部效果，上述技术方案中的一个技术方案具有如下优点或有益效果：The effects provided in the summary of the invention are only the effects of the embodiments, rather than all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:
1、通过对多源数据的获取及融合，提供多维度的电网薄弱环节指标，实时计算当前状态的指标值，并推送问题清单，采用深度学习算法预测预警，不断提高预警的准确性和预警结果的可靠性，对于问题清单和预警清单，提供闭环工作机制，统筹治理电网问题。1. Through the acquisition and integration of multi-source data, we provide multi-dimensional power grid weak link indicators, calculate the current status indicator values in real time, and push a list of problems. We use deep learning algorithms to predict early warnings and continuously improve the accuracy and results of early warnings. reliability, provide a closed-loop working mechanism for the problem list and early warning list, and coordinate the management of power grid problems.
2、电网信息感知模块融合内外部多源数据，诊断数据由人工收集变为自动获取，立足泛在电力物联网建设积累的数据资源，实现电网、客户、政府规划等信息的高度融合，实时更新电网内外部信息，为薄弱环节的诊断和治理提供数据基础。2. The power grid information sensing module integrates internal and external multi-source data, and the diagnostic data changes from manual collection to automatic acquisition. Based on the data resources accumulated in the construction of the ubiquitous power Internet of Things, it achieves a high degree of integration of power grid, customer, government planning and other information, and real-time updates Information inside and outside the power grid provides a data basis for the diagnosis and management of weak links.
3、电网现状诊断模块构建24维度诊断体系，薄弱环节由人工查找变为智能识别，深入分析电网薄弱环节异常特征，打造电网诊断神经感知体系，自主判定电网薄弱环节，逐站、逐线、逐台区对电网状态扫描和诊断评估，实时捕捉电网存在问题，智能生成问题清单。3. The power grid status diagnosis module builds a 24-dimensional diagnosis system, changing weak links from manual search to intelligent identification, in-depth analysis of the abnormal characteristics of power grid weak links, creating a power grid diagnosis neural perception system, and independently determining power grid weak links, station by station, line by line, and line by line. The Taiwan District scans and diagnoses the status of the power grid, captures problems in the power grid in real time, and intelligently generates a list of problems.
4、智能辅助决策模块建立全过程在线管理机制，解决方案由人工制定变为智能决策，智能提出备选解决措施，建立专业协同的全过程在线管理机制，实现方案优选与项目自动排序，提高电网投资的精准性。4. The intelligent auxiliary decision-making module establishes a whole-process online management mechanism. The solution is changed from manual formulation to intelligent decision-making. Alternative solutions are intelligently proposed, and a professional and collaborative whole-process online management mechanism is established to realize program selection and automatic project sequencing, and improve the power grid. Investment precision.
附图说明Description of the drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，对于本领域普通技术人员而言，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those of ordinary skill in the art, It is said that other drawings can also be obtained based on these drawings without exerting creative work.
图1是本发明所述系统的结构示意图；Figure 1 is a schematic structural diagram of the system according to the present invention;
图2是本发明所述方法的流程示意图；Figure 2 is a schematic flow chart of the method of the present invention;
图3是本发明进行电网信息感知及数据融合的架构图；Figure 3 is an architecture diagram of the present invention for power grid information sensing and data fusion;
图4是本发明电网现状诊断的原理示意图；Figure 4 is a schematic diagram of the principle of power grid status diagnosis according to the present invention;
图5是本发明24维度电网诊断神经感知体系的架构图；Figure 5 is an architectural diagram of the 24-dimensional power grid diagnostic neural sensing system of the present invention;
图6是本发明电网薄弱环节预警原理图；Figure 6 is a schematic diagram of early warning for weak links in the power grid of the present invention;
图7是本发明智能辅助决策阶段的原理图。Figure 7 is a schematic diagram of the intelligent assisted decision-making stage of the present invention.
具体实施方式Detailed ways
为能清楚说明本方案的技术特点，下面通过具体实施方式，并结合其附图，对本发明进行详细阐述。下文的公开提供了许多不同的实施例或例子用来实现本发明的不同结构。为了简化本发明的公开，下文中对特定例子的部件和设置进行描述。此外，本发明可以在不同例子中重复参考数字和/或字母。这种重复是为了简化和清楚的目的，其本身不指示所讨论各种实施例和/或设置之间的关系。应当注意，在附图中所图示的部件不一定按比例绘制。本发明省略了对公知组件和处理技术及工艺的描述以避免不必要地限制本发明。In order to clearly explain the technical features of this solution, the present invention will be described in detail below through specific implementation modes and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the invention. In order to simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Furthermore, the present invention may repeat reference numbers and/or letters in different examples. This repetition is for purposes of simplicity and clarity and does not by itself indicate a relationship between the various embodiments and/or arrangements discussed. It should be noted that components illustrated in the figures are not necessarily to scale. Descriptions of well-known components and processing techniques and processes are omitted to avoid unnecessarily limiting the invention.
如图1所示，本发明的电网薄弱环节动态监测预警决策系统包括电网信息感知模块、电网状态诊断模块、薄弱环节预警模块和智能辅助决策模块。As shown in Figure 1, the power grid weak link dynamic monitoring and early warning decision-making system of the present invention includes a power grid information sensing module, a power grid status diagnosis module, a weak link early warning module and an intelligent auxiliary decision-making module.
电网状态诊断模块建立电网诊断专家知识库，从多个维度定义电网薄弱环节指标，基于电网规划综合数据库，通过多算法引擎，实时计算所述电网薄弱环节的指标值，推送问题清单；薄弱环节预警模块基于所述电网规划综合数据库，建立参数档案和运行档案，结合影响因素，建立预警模型，采用人工智能或深度学习算法预测预警，形成预警清单；智能辅助决策模块对于问题清单和预警清单，提出解决方案，并向人工发送处理工单，收集人工处理进程及处理结果。The power grid status diagnosis module establishes a power grid diagnosis expert knowledge base, defines power grid weak link indicators from multiple dimensions, and based on the power grid planning comprehensive database, uses a multi-algorithm engine to calculate the indicator values of the power grid weak links in real time and push a problem list; weak link early warning Based on the power grid planning comprehensive database, the module establishes parameter files and operation files, combines influencing factors, establishes an early warning model, uses artificial intelligence or deep learning algorithms to predict early warnings, and forms an early warning list; the intelligent auxiliary decision-making module proposes a problem list and an early warning list. Solution, send processing work orders to manual, and collect manual processing progress and processing results.
电网信息感知包括多源数据接入、元数据存储、设备统一建模、模型匹配度计算和多远数据融合等功能单元；电网现状诊断包括电网诊断专家知识库、24维度诊断体系、问题清单和薄弱环节定位等功能单元；薄弱环节预警包括电网规划综合数据库、预警模型、机器学习计算和预警清单等功能单元；智能辅助决策包括诊断结论、解决措施推荐、专业协同优化、项目进展在线管理和闭环反馈等功能单元。Power grid information perception includes functional units such as multi-source data access, metadata storage, unified equipment modeling, model matching calculation and multi-remote data fusion; power grid status diagnosis includes power grid diagnosis expert knowledge base, 24-dimensional diagnosis system, problem list and Functional units such as weak link positioning; weak link early warning includes functional units such as power grid planning comprehensive database, early warning model, machine learning calculation and early warning list; intelligent auxiliary decision-making includes diagnostic conclusions, solution recommendation, professional collaborative optimization, online project progress management and closed loop Feedback and other functional units.
如图2所示，本发明的一种电网薄弱环节动态监测预警决策方法包括以下步骤：As shown in Figure 2, a dynamic monitoring and early warning decision-making method for weak links in the power grid of the present invention includes the following steps:
S1,通过ETL或服务调用的方式接入电力业务系统，获取业务系统数据，并通过网络爬虫获取用电量关联信息，形成电网规划综合数据库，并建立数据库内多源数据的关联关系，进行数据融合和电网设备建模；S1, access the power business system through ETL or service invocation, obtain business system data, and obtain electricity consumption related information through web crawlers, form a comprehensive database for power grid planning, and establish correlation relationships between multi-source data in the database to conduct data processing. convergence and grid equipment modeling;
S2,建立电网诊断专家知识库，从多个维度定义电网薄弱环节指标，基于所述电网规划综合数据库，通过多算法引擎，实时计算所述电网薄弱环节的指标值，推送问题清单；S2, establish a power grid diagnosis expert knowledge base, define power grid weak link indicators from multiple dimensions, calculate the indicator values of the power grid weak link in real time through a multi-algorithm engine based on the power grid planning comprehensive database, and push a problem list;
S3,基于所述电网规划综合数据库，建立参数档案和运行档案，结合影响因素，建立预警模型，采用人工智能或深度学习算法预测预警，形成预警清单；S3, based on the power grid planning comprehensive database, establish parameter files and operation files, combine influencing factors, establish an early warning model, use artificial intelligence or deep learning algorithms to predict early warnings, and form an early warning list;
S4,对于所述问题清单和预警清单，提出解决方案，并向人工发送处理工单，收集人工处理进程及处理结果。S4. For the problem list and early warning list, propose solutions, send manual processing work orders, and collect manual processing progress and processing results.
如图3所示，步骤S1中，多源数据接入包括PMS(power production managementsystem)是工程生产管理系统)系统、GIS(地理信息系统(Geographic Information System或Geo－Information system)系统、EMS(Energy Management System，能源管理系统)系统、供电服务指挥系统、用电信息采集系统等多个业务系统数据，通过网络爬虫技术获取相关政府网站经济、人口、能源、政府规划等用电量关联信息，形成集电网设备台账、设备拓扑关系、设备实时运行信息、设备历史运行数据、设备空间地理信息、经济、人口、能源、政府规划于一体的电网规划综合数据库。As shown in Figure 3, in step S1, multi-source data access includes PMS (power production management system) system, GIS (Geographic Information System or Geo-Information system) system, EMS (Energy Management System (Energy Management System) system, power supply service command system, electricity consumption information collection system and other business system data, through web crawler technology to obtain relevant government website economy, population, energy, government planning and other electricity consumption related information to form A comprehensive database for power grid planning that integrates power grid equipment ledgers, equipment topological relationships, equipment real-time operation information, equipment historical operation data, equipment spatial geographical information, economy, population, energy, and government planning.
形成的电网规划综合数据库包括设备数据库、运行数据库、图形数据库和规划数据库。The formed comprehensive power grid planning database includes equipment database, operation database, graphics database and planning database.
对源数据的存储包括数据结构的存储和数据处理。数据结构的存储包括结构化数据、非结构化数据、空间数据、海量历史数据和准实时数据；数据处理包括数据抽取、数据清洗、更新机制和大数据储存。The storage of source data includes the storage of data structures and data processing. The storage of data structures includes structured data, unstructured data, spatial data, massive historical data and quasi-real-time data; data processing includes data extraction, data cleaning, update mechanisms and big data storage.
由于多源系统数据相对独立，缺少关联关系，无法直接使用，需通过以下方式实现多源异构数据融合：Since multi-source system data is relatively independent and lacks correlation, it cannot be used directly. Multi-source heterogeneous data fusion needs to be achieved in the following ways:
建立包括各源系统数据逻辑关系、拓扑结构、空间信息、特征属性等模型；逻辑模型包括树形结构、设备类型和电压等级；拓扑模型包括有向图结构、电气连接拓扑和空间连接拓扑；特征模型包括线性结构、特征属性提取和空间属性提取。Establish models including logical relationships, topological structures, spatial information, feature attributes, etc. of data from each source system; logical models include tree structures, equipment types, and voltage levels; topological models include directed graph structures, electrical connection topology, and spatial connection topology; features The model includes linear structure, feature attribute extraction and spatial attribute extraction.
通过模型各属性信息遍历比对，计算模型间匹配度；模型间匹配度的计算依次通过根节点对应、逻辑模型遍历、有向图遍历、特征属性识别、空间坐标转换和图像识别的方式进行。Through the traversal and comparison of each attribute information of the model, the matching degree between models is calculated; the matching degree between models is calculated through root node correspondence, logical model traversal, directed graph traversal, feature attribute identification, spatial coordinate conversion and image recognition.
根据匹配度情况进行自动关联，实现数据自动集成融合；将匹配结果分为三类，用A、B和C标识。匹配度-A是指自动关联设备信息，并记录数据在各系统间的细微差异日志；匹配度-B是指人工再次确认后，通过系统功能融合，并记录数据匹配依据及未自动匹配原因；匹配度-C是指属缺失匹配数据，记录原因，通知元数据系统进行数据修补。对未能融合数据按数据所属电压等级和设备所在区域以任务的方式在系统中分发给相应用户进行手工修补。Automatic association is performed based on the matching degree to achieve automatic data integration and fusion; the matching results are divided into three categories, identified by A, B and C. Matching degree-A refers to automatically correlating device information and recording the log of subtle differences in data between systems; matching degree-B refers to merging the data through system functions after manual reconfirmation, and recording the data matching basis and the reasons for non-automatic matching; Matching degree -C refers to missing matching data, record the reason, and notify the metadata system to perform data repair. The data that cannot be fused is distributed to the corresponding users in the system in the form of tasks according to the voltage level of the data and the area where the equipment is located for manual repair.
建立变电站、变压器、线路等电网设备模型，模型包括容量、长度、投运时间等设备参数，设备N-1通过率、重过载次数、重过载时长、容载比、故障次数等指标，利用数据挖掘、图像识别、矢量坐标转换等方法，可视化全景展示电网信息，全面检查各项指标，解决精准感知难的问题。Establish models of power grid equipment such as substations, transformers, and lines. The model includes equipment parameters such as capacity, length, and operation time, as well as indicators such as equipment N-1 pass rate, number of heavy overloads, duration of heavy overloads, capacity-to-load ratio, and number of failures. Utilize data Mining, image recognition, vector coordinate conversion and other methods can be used to visually display power grid information in a panoramic manner, comprehensively inspect various indicators, and solve the problem of difficulty in accurate perception.
如图4、5所示，步骤S2中，根据电网相关标准、历年事故案例、电网经济性要求、电网薄弱环节分类构建电网诊断专家知识库，根据专家知识库内容，从供电能力、电网结构、设备水平、电压质量、效率效益5个维度，定义重过载、线路多T接、老旧设备、多站串供等24个关键薄弱环节指标，依据电网薄弱环节指标判定标准(如线路重过载定义为线路负载率达到70％即为重载，负载率超过100％即为过载)，定义指标计算公式，设置指标阈值，利用分布式并行计算和数据探针技术，深入综合数据库内部，通过拓扑识别、N-1计算、负载分析、容载比计算、供电范围识别、运行方式识别等算法引擎，依据24维度诊断体系及相关计算规则，实时计算指标得分，动态监测电网设备运行状态，实时推送设备问题清单，问题清单包括：设备信息、问题成因、指标得分、空间位置、问题紧迫程度。As shown in Figures 4 and 5, in step S2, a power grid diagnosis expert knowledge base is constructed based on power grid related standards, accident cases over the years, power grid economic requirements, and power grid weak link classification. According to the content of the expert knowledge base, power supply capacity, power grid structure, The five dimensions of equipment level, voltage quality, and efficiency benefit define 24 key weak link indicators such as heavy overload, multiple T-connections on lines, old equipment, and multi-station series supply. Based on the power grid weak link indicator determination standards (for example, heavy overload on lines is defined as When the line load rate reaches 70%, it is overloaded, and when the load rate exceeds 100%, it is overloaded). Define the index calculation formula, set the index threshold, use distributed parallel computing and data probe technology, go deep into the comprehensive database, and through topology identification, N-1 calculation, load analysis, capacity-to-load ratio calculation, power supply range identification, operating mode identification and other algorithm engines, based on the 24-dimensional diagnosis system and related calculation rules, calculate index scores in real time, dynamically monitor the operating status of power grid equipment, and push equipment problems in real time List, the problem list includes: equipment information, problem cause, indicator score, spatial location, and problem urgency.
24维度电网诊断神经感知体系分为高压配电网的诊断体系和中、低压配电网的诊断体系。高压配电网分别对供电能力、网架结构、设备水平和效率效益进行诊断，具体包括供电能力方面：局部供电能力不足、主变重过载和线路重过载；网架结构方面：单线单边站、同塔双回且双辐射供电、单条线路多T接、线路不满足N-1校验、主变不满足N-1校验、35KV变电站高压侧进线30°相角差；设备水平方面：老旧变电站和老旧线路；效率效益方面：主变轻载和线路轻载。The 24-dimensional power grid diagnostic neural sensing system is divided into a diagnostic system for high-voltage distribution networks and a diagnostic system for medium- and low-voltage distribution networks. The high-voltage distribution network diagnoses the power supply capacity, grid structure, equipment level and efficiency benefits respectively, including power supply capacity: insufficient local power supply capacity, main transformer heavy overload and line heavy overload; grid structure: single line single side station , double circuit and double radiation power supply on the same tower, multiple T connections on a single line, the line does not meet the N-1 verification, the main transformer does not meet the N-1 verification, the high-voltage side of the 35KV substation has a 30° phase angle difference; equipment level : Old substations and old lines; In terms of efficiency and benefits: light load of main transformer and light load of lines.
中、低压配电网分别对供电能力、网架结构、设备水平、效率效益和电压质量进行诊断。供电能力方面：配变重过载和线路重过载；网架结构方面：单辐射线路和线路不满足N-1校验；设备水平方面：老旧配变、老旧线路和10KV架空线路绝缘化水平；效率效益方面：配变轻载和线路轻载；电压质量方面包括电压质量问题。The medium and low-voltage distribution networks diagnose the power supply capacity, grid structure, equipment level, efficiency and benefit, and voltage quality respectively. In terms of power supply capacity: heavy overload of distribution transformers and heavy overload of lines; in terms of grid structure: single radiation lines and lines do not meet N-1 verification; in terms of equipment level: insulation level of old distribution transformers, old lines and 10KV overhead lines ; In terms of efficiency and benefit: light load of distribution transformers and light loads of lines; voltage quality includes voltage quality issues.
如图6所示，步骤S3中，构建综合历史数据与多影响因素的预警模型，实现电网运行状态的超前预判及潜在风险的及时防控。基于综合数据库，逐站、逐线、逐台区建立参数档案和运行档案，结合负荷曲线、新增报装、电网建设、供电裕度、用户性质和气象信息等影响因素，建立预警模型，采用人工智能和/或深度学习算法预测预警，形成预警清单，专业人员对照清单给予重点关注，优先立项，做到追根溯源、精准预判。As shown in Figure 6, in step S3, an early warning model that integrates historical data and multiple influencing factors is constructed to achieve advance prediction of the operating status of the power grid and timely prevention and control of potential risks. Based on the comprehensive database, parameter files and operation files are established station by station, line by line, and zone by zone. An early warning model is established based on influencing factors such as load curves, new installations, power grid construction, power supply margin, user nature, and meteorological information. Artificial intelligence and/or deep learning algorithms predict early warnings and form an early warning list. Professionals will pay special attention to the list and prioritize projects to achieve traceability and accurate prediction.
应用机器学习技术逐步提升预警精准度，以设备重过载预警为例，先设置一个经验参数，针对历史负载率数据逐月预测，将计算结果与实际发生结果进行拟合度比对，然后对每个参数按百分位精度进行调整，逐步提高预警精度，并利用调整结果进行问题预警。Apply machine learning technology to gradually improve the accuracy of early warning. Taking the equipment heavy overload warning as an example, first set an empirical parameter, predict the historical load rate data month by month, compare the calculation results with the actual occurrence results, and then compare each Each parameter is adjusted according to percentile accuracy, gradually improving the early warning accuracy, and using the adjustment results to provide early warning of problems.
如图7所示，基于诊断问题清单和预警清单，建立包括发展专业、运维专业、调控专业、基建专业和营销专业在内的各专业协同的电网诊断和问题治理全过程在线管理机制，实现科学有序治理电网薄弱环节。对于诊断出的问题和预警清单，利用机器学习技术提出备选解决方案，发起处理工单，实时传输至相关专业人员，专业人员通过移动终端接收工单，优选完善解决方案，并将进展情况实时上传至系统中，建立诊断、预警、方案优选和处理反馈的闭环工作机制，利用薄弱环节核查结果反复优化分析模型，不断提高智能诊断和辅助决策的准确性。项目进展情况的在线管理贯穿于规划阶段、可研阶段、投资计划阶段、项目建设阶段和项目后评价阶段。智能辅助决策的引入，提升专业协同，规划统筹制定解决方案，实现以最小的投入科学有序治理薄弱环节。建立各专业协调顺畅的闭环工作机制，做到以规划为引领，统筹治理电网问题。As shown in Figure 7, based on the diagnostic problem list and early warning list, an online management mechanism for the entire process of power grid diagnosis and problem management is established, including development majors, operation and maintenance majors, regulation majors, infrastructure majors, and marketing majors, to achieve Manage weak links in the power grid in a scientific and orderly manner. For diagnosed problems and warning lists, machine learning technology is used to propose alternative solutions, initiate processing work orders, and transmit them to relevant professionals in real time. Professionals receive work orders through mobile terminals, optimize and improve solutions, and report progress in real time. Upload to the system, establish a closed-loop working mechanism for diagnosis, early warning, solution optimization and processing feedback, use the weak link verification results to repeatedly optimize the analysis model, and continuously improve the accuracy of intelligent diagnosis and assisted decision-making. The online management of project progress runs through the planning stage, feasibility study stage, investment planning stage, project construction stage and post-project evaluation stage. The introduction of intelligent assisted decision-making improves professional collaboration, plans and coordinates the formulation of solutions, and achieves scientific and orderly management of weak links with minimal investment. Establish a coordinated and smooth closed-loop working mechanism for various disciplines to coordinate and manage power grid issues with planning as the guide.
以上所述只是本发明的优选实施方式，对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也被视为本发明的保护范围。The above are only preferred embodiments of the present invention. For those of ordinary skill in the art, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also regarded as the present invention. protection scope of the invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911191417.XACN111143447B (en) | 2019-11-28 | 2019-11-28 | A dynamic monitoring and early warning decision-making system and method for weak links in the power grid |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911191417.XACN111143447B (en) | 2019-11-28 | 2019-11-28 | A dynamic monitoring and early warning decision-making system and method for weak links in the power grid |
| Publication Number | Publication Date |
|---|---|
| CN111143447A CN111143447A (en) | 2020-05-12 |
| CN111143447Btrue CN111143447B (en) | 2023-10-31 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911191417.XAActiveCN111143447B (en) | 2019-11-28 | 2019-11-28 | A dynamic monitoring and early warning decision-making system and method for weak links in the power grid |
| Country | Link |
|---|---|
| CN (1) | CN111143447B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111817265B (en)* | 2020-06-29 | 2023-03-21 | 北京智芯微电子科技有限公司 | Low-voltage transformer area power distribution protection method and system |
| CN112907081B (en)* | 2021-02-22 | 2022-09-06 | 国网安徽省电力有限公司合肥供电公司 | Method and system for intelligently diagnosing operation problems of agricultural distribution network |
| CN113685335B (en)* | 2021-08-27 | 2023-07-25 | 广东鑫钻节能科技股份有限公司 | Safety energy-saving air compression station with electric energy quality monitoring system |
| CN113723869A (en)* | 2021-11-02 | 2021-11-30 | 国网北京市电力公司 | Power grid infrastructure project post-evaluation method, device, equipment and medium |
| CN114418306B (en)* | 2021-12-21 | 2025-09-09 | 南方电网数据平台与安全(广东)有限公司 | Artificial intelligence review system based on NLP technology |
| CN114860839A (en)* | 2022-07-07 | 2022-08-05 | 中国长江三峡集团有限公司 | Power distribution network data analysis and visualization system and method |
| CN117707027B (en)* | 2024-02-05 | 2024-04-16 | 国网甘肃省电力公司白银供电公司 | A one-button sequential control system for centralized control stations based on multi-source data fusion |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101179195A (en)* | 2007-11-15 | 2008-05-14 | 上海交通大学 | Auxiliary decision-making system for distribution network planning scheme |
| CN102708472A (en)* | 2012-06-20 | 2012-10-03 | 江西省电力公司超高压分公司 | Intelligent processing method for warning information of combined power grid |
| CN103647276A (en)* | 2013-12-10 | 2014-03-19 | 国家电网公司 | Electric energy quality early warning system and method thereof |
| CN103679293A (en)* | 2013-12-20 | 2014-03-26 | 长园深瑞继保自动化有限公司 | Intelligent substation warning and aid decision making system |
| CN106026383A (en)* | 2016-05-16 | 2016-10-12 | 国网辽宁省电力有限公司朝阳供电公司 | Intelligent transformer station equipment state multi-system cooperative diagnosis method based on information fusion |
| CN107274115A (en)* | 2017-08-11 | 2017-10-20 | 国网江苏省电力公司电力科学研究院 | Active distribution network Situation Awareness System and method based on distributed monitoring and Multi-source Information Fusion |
| CN107402976A (en)* | 2017-07-03 | 2017-11-28 | 国网山东省电力公司经济技术研究院 | Power grid multi-source data fusion method and system based on multi-element heterogeneous model |
| CN107464062A (en)* | 2017-08-09 | 2017-12-12 | 国网山东省电力公司经济技术研究院 | A kind of power network development Evaluation of Diagnosis system |
| CN109343995A (en)* | 2018-10-25 | 2019-02-15 | 金税信息技术服务股份有限公司 | Intelligent O&M analysis system based on multi-source heterogeneous data fusion, machine learning and customer service robot |
| CN109816161A (en)* | 2019-01-14 | 2019-05-28 | 中国电力科学研究院有限公司 | A kind of distribution network operation auxiliary decision analysis system and its application method |
| CN109978295A (en)* | 2017-12-28 | 2019-07-05 | 南京易司拓电力科技股份有限公司 | A kind of lean decision system of data-driven power distribution network |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101179195A (en)* | 2007-11-15 | 2008-05-14 | 上海交通大学 | Auxiliary decision-making system for distribution network planning scheme |
| CN102708472A (en)* | 2012-06-20 | 2012-10-03 | 江西省电力公司超高压分公司 | Intelligent processing method for warning information of combined power grid |
| CN103647276A (en)* | 2013-12-10 | 2014-03-19 | 国家电网公司 | Electric energy quality early warning system and method thereof |
| CN103679293A (en)* | 2013-12-20 | 2014-03-26 | 长园深瑞继保自动化有限公司 | Intelligent substation warning and aid decision making system |
| CN106026383A (en)* | 2016-05-16 | 2016-10-12 | 国网辽宁省电力有限公司朝阳供电公司 | Intelligent transformer station equipment state multi-system cooperative diagnosis method based on information fusion |
| CN107402976A (en)* | 2017-07-03 | 2017-11-28 | 国网山东省电力公司经济技术研究院 | Power grid multi-source data fusion method and system based on multi-element heterogeneous model |
| CN107464062A (en)* | 2017-08-09 | 2017-12-12 | 国网山东省电力公司经济技术研究院 | A kind of power network development Evaluation of Diagnosis system |
| CN107274115A (en)* | 2017-08-11 | 2017-10-20 | 国网江苏省电力公司电力科学研究院 | Active distribution network Situation Awareness System and method based on distributed monitoring and Multi-source Information Fusion |
| CN109978295A (en)* | 2017-12-28 | 2019-07-05 | 南京易司拓电力科技股份有限公司 | A kind of lean decision system of data-driven power distribution network |
| CN109343995A (en)* | 2018-10-25 | 2019-02-15 | 金税信息技术服务股份有限公司 | Intelligent O&M analysis system based on multi-source heterogeneous data fusion, machine learning and customer service robot |
| CN109816161A (en)* | 2019-01-14 | 2019-05-28 | 中国电力科学研究院有限公司 | A kind of distribution network operation auxiliary decision analysis system and its application method |
| Publication number | Publication date |
|---|---|
| CN111143447A (en) | 2020-05-12 |
| Publication | Publication Date | Title |
|---|---|---|
| CN111143447B (en) | A dynamic monitoring and early warning decision-making system and method for weak links in the power grid | |
| CN109816161B (en) | A distribution network operation auxiliary decision analysis system and its application method | |
| CN113159339B (en) | One-area one-index line loss management method and system based on big data | |
| CN116865258B (en) | A hierarchical structure distributed power smart grid construction method | |
| CN108336725B (en) | Power grid dispatching equipment monitoring management and intelligent analysis system | |
| CN107133255B (en) | Panoramic security defense method and system for large power grid | |
| CN118739184A (en) | A method for automatic fault detection and repair of self-healing intelligent power lines | |
| CN115293507A (en) | Situation knowledge graph construction method and system for power grid dispatching operation management | |
| CN106649840A (en) | Method suitable for power data quality assessment and rule check | |
| CN118487276B (en) | A method and system for dynamic control of power grid security for power security objects | |
| CN103902816A (en) | Electrification detection data processing method based on data mining technology | |
| CN105787606A (en) | Power dispatching online trend early warning system based on ultra short term load prediction | |
| CN102509178A (en) | Distribution network device status evaluating system | |
| CN104331844A (en) | Power network infrastructure project investment decision-making method | |
| CN115511656A (en) | An Auxiliary Decision-Making System for Demand Planning Based on Mining the Value of Grid Data | |
| Zhao | Research on management informatization construction of electric power enterprise based on big data technology | |
| CN118611060B (en) | Control strategy generation method, system, device and storage medium of power supply device | |
| CN113949155A (en) | Panoramic power quality monitoring system with real-time monitoring function | |
| CN113837423A (en) | Power grid operation situation prediction method based on energy internet electric power big data | |
| CN118504393A (en) | Autoformer prediction-based power distribution network digital twin body construction method and system | |
| CN118735118A (en) | Power supply service construction system and method in different scenarios based on digital twin | |
| CN112053086B (en) | Working method of operation diagnosis system based on big data power grid | |
| CN112183895B (en) | Intelligent identification and auxiliary compilation system for regional power grid maintenance plan | |
| CN113269435A (en) | New energy station running state coupling monitoring and evaluating system | |
| CN115983548A (en) | An intelligent planning method for distribution network based on artificial intelligence technology |
| 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 |