CN115344717B - Method and device for constructing knowledge map for multi-type energy supply and utilization system regulation and operation - Google Patents
Method and device for constructing knowledge map for multi-type energy supply and utilization system regulation and operationDownload PDFInfo
- Publication number
- CN115344717B CN115344717BCN202211272027.7ACN202211272027ACN115344717BCN 115344717 BCN115344717 BCN 115344717BCN 202211272027 ACN202211272027 ACN 202211272027ACN 115344717 BCN115344717 BCN 115344717B
- Authority
- CN
- China
- Prior art keywords
- data
- knowledge
- knowledge map
- layer
- energy supply
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Translated fromChinese
Description
Translated fromChinese技术领域technical field
本发明属于能源调度技术领域，具体涉及一种面向多类型供用能系统调控运行知识图谱构建方法及装置。The invention belongs to the technical field of energy dispatching, and in particular relates to a method and device for constructing a knowledge map for regulation and operation of multi-type energy supply and utilization systems.
背景技术Background technique
大规模新能源发电接入电网，其具有间歇性、随机性和波动性，使电网呈现出更加复杂的非线性随机特性、多状态变量耦合及多时间尺度动态特性。同时也使得系统形态及运行特性日益复杂，配电网交直流之间、直流送受端电网之间耦合关系复杂，网络结构变化大、运行方式复杂多变，监控断面数量日益增加，控制规则日趋复杂。这些对电网稳定运行提出严峻挑战，对电力调度运行提出更精细化要求。The large-scale new energy generation connected to the power grid is intermittent, random and volatile, which makes the power grid present more complex nonlinear random characteristics, multi-state variable coupling and multi-time scale dynamic characteristics. At the same time, the system form and operation characteristics are becoming more and more complex, the coupling relationship between the AC and DC of the distribution network, and the DC transmission and receiving end grid is complex, the network structure changes greatly, the operation mode is complex and changeable, the number of monitoring sections is increasing, and the control rules are becoming more and more complicated. . These pose serious challenges to the stable operation of the power grid, and put forward more refined requirements for power dispatching operations.
因此电网调度运行自动化系统对电网的安全运行越来越起着不可或缺的作用。现有调度自动化系统各类应用越来越多，业务知识越来越复杂，相关业务人员大多只了解局部的业务知识，而不清楚与之相关的业务逻辑，只有极少数专家级人员才能对整个业务流程有清晰的认识。因此当复杂的业务逻辑出现问题时，需要临时调动所有业务人员，理清业务逻辑关系，才有可能找出引起问题的原因。而若建立起整个调度自动化系统相关业务的知识图谱，就可以根据知识图谱所表达的相关业务逻辑，十分清晰地查询所有的操作和数据流，从而找出最优的控制运行策略。Therefore, the dispatching operation automation system of the power grid plays an increasingly indispensable role in the safe operation of the power grid. There are more and more various applications in the existing scheduling automation system, and the business knowledge is becoming more and more complex. Most of the relevant business personnel only understand partial business knowledge, but do not know the related business logic. Business process has a clear understanding. Therefore, when there is a problem with the complex business logic, it is necessary to temporarily mobilize all business personnel to clarify the business logic relationship, so that it is possible to find out the cause of the problem. However, if a knowledge map of related businesses of the entire scheduling automation system is established, all operations and data flows can be queried very clearly according to the relevant business logic expressed in the knowledge map, so as to find the optimal control operation strategy.
因此，需要对各类文本形式的调控内容、运行策略等进行数字化处理，已有的相关研究主要通过采用计算机脚本语言对稳定规定规则进行逐条翻译，将形成的稳定运行控制策略计算结果，存储于 D5000 实时库，用于在线运行监控。比如在文章“基于智能分析的电网运行方式和安控策略智能编制方法和应用”中参考 C 语言定义方式，用脚本文件描述电网运行控制条件和手段内容及其之间的关联关系，以及安全自动装置动作条件和动作策略内容及其之间的关联关系；但是该方法逻辑关系不够直观、准确率不可控、脚本化录入和日常维护十分复杂，工作量巨大，同时脚本文档的形成和维护工作需要同时具备电力系统专业知识和编程技能，十分具有挑战性。知识图谱可以有效组织、管理和利用海量信息，实现智能化的知识抽取、推理、存储和检索，其特性及应用场景与电力系统的需要十分契合因此在电力系统自动化领域，针对运行控制自动化系统的相关业务逻辑，提出构建知识图谱的方法，建立自动化系统的知识图谱，便于调控策略的总结、搜索与应用，具有很高的研究价值。Therefore, it is necessary to digitize the regulatory content and operating strategies in various text forms. Existing related research mainly uses computer script language to translate the stability regulations one by one, and stores the calculation results of the stable operation control strategy in the D5000 real-time library for online operation monitoring. For example, in the article "Power grid operation mode based on intelligent analysis and intelligent compilation method and application of security control strategy", refer to the C language definition method, use script files to describe the power grid operation control conditions and means, and the relationship between them, as well as security and automatic control. Device action conditions and action strategy content and the relationship between them; however, the logical relationship of this method is not intuitive enough, the accuracy is uncontrollable, the script input and daily maintenance are very complicated, and the workload is huge. At the same time, the formation and maintenance of script documents require Combining power system expertise with programming skills is challenging. Knowledge graphs can effectively organize, manage and utilize massive amounts of information, and realize intelligent knowledge extraction, reasoning, storage and retrieval. Its characteristics and application scenarios are very suitable for the needs of power systems. Related business logic, the method of constructing knowledge map is proposed, and the knowledge map of automation system is established to facilitate the summary, search and application of control strategies, which has high research value.
发明内容Contents of the invention
本发明要解决的技术问题是针对上述现有技术的不足，提供面向多类型供用能系统调控运行知识图谱构建方法及装置，通过确立知识图谱模式层、构建不同的神经网络模型以实现调控运行策略数据的知识抽取并得到包含实体-属性-实体的三元组数据，并基于三元组数据构建知识图谱数据层，实现了电网调控策略文本的快速结构化提取和知识图谱构建及入库处理，节省了大量人工提取入库时间。The technical problem to be solved by the present invention is to provide a method and device for constructing a knowledge graph for the control and operation of multi-type energy supply and utilization systems in view of the above-mentioned deficiencies in the prior art, and realize the control and operation strategy by establishing the knowledge map model layer and constructing different neural network models The knowledge extraction of the data and the triple data containing entity-attribute-entity are obtained, and the knowledge map data layer is constructed based on the triple data, which realizes the fast structured extraction of the power grid control strategy text, the construction of the knowledge map and the storage processing. Save a lot of time for manual extraction and warehousing.
为解决上述技术问题，本发明所采取的技术方案是：In order to solve the problems of the technologies described above, the technical solution adopted in the present invention is:
一种面向多类型供用能系统调控运行知识图谱构建方法，包括以下步骤：A method for constructing a knowledge graph for the regulation and operation of multi-type energy supply and consumption systems, comprising the following steps:
步骤1：确定多类型供用能系统调控运行策略的运行技术指标，并依次细化，采用自顶向下的方法构建知识图谱模式层；Step 1: Determine the operation technical indicators of the multi-type energy supply and use system regulation and operation strategy, and refine them in turn, and use the top-down method to build the knowledge map model layer;
步骤2：对调控运行策略数据进行数据预处理；Step 2: Perform data preprocessing on the control and operation strategy data;
步骤3：构建不同的神经网络模型以实现调控运行策略数据的知识抽取并得到包含“实体-属性-实体”的三元组数据；Step 3: Construct different neural network models to realize the knowledge extraction of control operation strategy data and obtain the triple data containing "entity-attribute-entity";
步骤4：将三元组数据加入知识图谱中并采用自底向上的方法构建知识图谱数据层；Step 4: Add the triplet data to the knowledge map and use the bottom-up method to construct the knowledge map data layer;
步骤5：将知识图谱数据层数据进行归类并与知识图谱模式层结合形成策略集，构建出完整的知识图谱并存储。Step 5: Classify the knowledge map data layer data and combine it with the knowledge map pattern layer to form a policy set, construct a complete knowledge map and store it.
进一步优选，步骤1中，依据多类型供用能系统配电网实际运行状态和调控需求，确定运行技术指标为资源类型、组合方式、用户组成、运行指标；并向下细化，以完成知识图谱模式层的构建。Further preferably, in step 1, according to the actual operating status and control requirements of the multi-type energy supply and consumption system distribution network, determine the operating technical indicators as resource type, combination mode, user composition, and operating indicators; and refine downwards to complete the knowledge map Construction of the schema layer.
进一步优选，数据预处理包括对调控运行策略数据进行清洗、删除与补充，将调控运行策略数据按照特征指标划分为训练样本与测试样本；对调控运行策略数据进行分词，以得到调控词库，并根据调控词库，构建空间向量。Further preferably, the data preprocessing includes cleaning, deleting and supplementing the control operation strategy data, dividing the control operation strategy data into training samples and test samples according to characteristic indicators; performing word segmentation on the control operation strategy data to obtain a control lexicon, and According to the regulatory lexicon, construct the space vector.
进一步优选，分词和构建空间向量的过程为：首先确定查询文本预处理之后所含的每一个文档，并根据文档所属领域词库，对每一个文档进行分词，将每一个文档中的每个所属的备选关键词代入所得到的分析模型中，得到每一个文档中各备选关键词对应形成的若干维度的词向量，并作为空间向量输出。Further preferably, the process of word segmentation and construction of space vectors is as follows: firstly determine each document included in the query text after preprocessing, and perform word segmentation on each document according to the thesaurus of the field to which the document belongs, and classify each document in each document to Substitute the candidate keywords into the obtained analysis model, and obtain the word vectors of several dimensions corresponding to each candidate keyword in each document, and output them as space vectors.
步骤3的过程包括：Step 3 process includes:
步骤3.1：基于字向量使用TextCNN模型对调控运行策略数据进行文本分类；Step 3.1: Use the TextCNN model to classify the text of the control operation strategy data based on the word vector;
步骤3.2：使用LR-CNN模型实现对调控运行策略数据的命名实体识别；Step 3.2: Use the LR-CNN model to realize the named entity recognition of the control operation strategy data;
步骤3.3：运用BERT-BiLSTM-CRF模型对调控运行策略数据中的实体进行实体间关系抽取。Step 3.3: Use the BERT-BiLSTM-CRF model to extract the relationship between entities in the control operation strategy data.
进一步优选，步骤3.2中所述命名实体识别是对调控运行策略数据文本中具有特定含义的实体进行边界确定和类别识别，根据乡村配电网中包含的实体属性与种类，使用LR-CNN模型将其中实体划分为设备名称、资源名称、投资金额、回收周期、运行地点。Further preferably, the named entity recognition described in step 3.2 is to perform boundary determination and category recognition on entities with specific meanings in the control and operation strategy data text, and use the LR-CNN model to The entities are divided into equipment name, resource name, investment amount, payback period, and operating location.
进一步优选，所述 LR-CNN模型用CNN层对句子的字符、候选词特征进行提取，并使用注意力机制模块合并字符和候选词信息；同时引入 Rethinking 机制，向每个 CNN 层添加反馈层，使用高层的字词信息来调整低层的注意力机制模块的权重。Further preferably, the LR-CNN model uses the CNN layer to extract the characters and candidate word features of the sentence, and uses the attention mechanism module to merge the character and candidate word information; while introducing a Rethinking mechanism, adding a feedback layer to each CNN layer, Use the high-level word information to adjust the weight of the low-level attention mechanism module.
进一步优选，步骤3.3所述实体间关系抽取是在命名实体识别的基础上判断实体间是否存在预定义的关系，从而构成一系列三元组知识；通过配电网运行规范与状态对实体间关系进行预定义，利用抽取后形成的关系构建知识图谱中的一系列三元组。Further preferably, the entity-to-entity relationship extraction described in step 3.3 is to judge whether there is a predefined relationship between entities on the basis of named entity recognition, thereby forming a series of triplet knowledge; Predefine and use the relationship formed after extraction to construct a series of triples in the knowledge graph.
进一步优选，所述BERT-BiLSTM-CRF模型使用预训练过的 BERT 模型，在BERT 模型后面连接 BiLSTM-CRF 模型，同时将 BERT 模型的输出和 BiLSTM 层的输出进行特征串联，并对其进行微调训练，进一步提高关系抽取的准确率。Further preferably, the BERT-BiLSTM-CRF model uses a pre-trained BERT model, and the BiLSTM-CRF model is connected behind the BERT model, and at the same time, the output of the BERT model and the output of the BiLSTM layer are subjected to feature concatenation, and fine-tuning training is performed on it , to further improve the accuracy of relation extraction.
进一步优选，所述步骤5具体为：Further preferably, the step 5 is specifically:
步骤5.1：基于多类型供用能系统配电网资源种类及特性确定典型场景；Step 5.1: Determine typical scenarios based on the types and characteristics of distribution network resources in multi-type energy supply and consumption systems;
步骤5.2：将知识图谱数据层中的调控要求整理为策略集；Step 5.2: Organize the regulatory requirements in the knowledge map data layer into a policy set;
步骤5.3：将知识图谱数据层中的具体数据归纳为策略并放入策略集；Step 5.3: Summarize the specific data in the knowledge map data layer into strategies and put them into the strategy set;
步骤5.4：将知识图谱数据层与知识图谱模式层结合形成完整的知识图谱；Step 5.4: Combine the knowledge map data layer and the knowledge map pattern layer to form a complete knowledge map;
步骤5.5：存储知识图谱。Step 5.5: Store the knowledge graph.
本发明提供一种面向多类型供用能系统调控运行知识图谱构建方法的装置，包括知识图谱模式层模块、数据预处理模块、知识抽取模块、知识图谱数据层模块、策略集模块和存储模块；The present invention provides a device oriented to multi-type energy supply and utilization system regulation and operation knowledge map construction method, including a knowledge map model layer module, a data preprocessing module, a knowledge extraction module, a knowledge map data layer module, a policy set module and a storage module;
所述知识图谱模式层模块用于确定多类型供用能系统调控运行策略的运行技术指标，并依次细化形成知识图谱模式层；The knowledge map model layer module is used to determine the operation technical indicators of the multi-type energy supply and utilization system regulation and control operation strategy, and successively refine the knowledge map model layer;
所述数据预处理模块用于对调控运行策略数据进行数据预处理；The data preprocessing module is used to perform data preprocessing on the regulation and control operation strategy data;
所述知识抽取模块用于构建不同的神经网络模型以实现调控运行策略数据的知识抽取并得到包含实体-属性-实体的三元组数据；The knowledge extraction module is used to construct different neural network models to realize the knowledge extraction of control operation strategy data and obtain triple data comprising entity-attribute-entity;
所述知识图谱数据层模块将三元组数据加入知识图谱中并采用自底向上的方法构建知识图谱数据层；The knowledge graph data layer module adds triplet data into the knowledge graph and adopts a bottom-up method to construct the knowledge graph data layer;
所述策略集模块将知识图谱数据层数据进行归类并与知识图谱模式层结合形成策略集；所述存储模块用于存储知识图谱。The strategy set module classifies the knowledge graph data layer data and combines it with the knowledge graph pattern layer to form a strategy set; the storage module is used to store the knowledge graph.
采用上述技术方案所产生的有益效果在于：本发明基于深度学习和知识图谱领域的相关技术，通过确立知识图谱模式层、构建不同的神经网络模型以实现调控运行策略数据的知识抽取并得到包含实体-属性-实体的三元组数据，并基于三元组数据构建知识图谱数据层，实现了电网调控策略文本的快速结构化提取和知识图谱构建及入库处理，节省了大量人工提取入库时间。使用本发明的知识图谱进行调控信息检索时，不但能对相应调控策略进行快速、准确信息定位，实现检索内容推送，并且能够提供具体的操作信息 ，方便了工作人员的查找与处置，且准确性较高，克服了现有知识图谱的局限性；同时，电网调控决策知识图谱可进行多层次、多阶段决策支持信息推送，提高了数据查询检索效率。The beneficial effects produced by adopting the above technical solution are: the present invention is based on the related technologies in the field of deep learning and knowledge graph, by establishing the knowledge graph model layer and constructing different neural network models to realize the knowledge extraction of control operation strategy data and obtain the contained entities -Triple data of attribute-entity, and build a knowledge map data layer based on the triple data, which realizes the rapid structured extraction of power grid control strategy text and knowledge map construction and storage processing, saving a lot of time for manual extraction and storage . When the knowledge map of the present invention is used for regulatory information retrieval, not only can the corresponding regulatory strategies be quickly and accurately located, and the retrieval content can be pushed, but also can provide specific operation information, which facilitates the search and disposal of the staff, and is accurate. Higher, overcoming the limitations of the existing knowledge map; at the same time, the power grid regulation and decision-making knowledge map can carry out multi-level and multi-stage decision support information push, which improves the efficiency of data query and retrieval.
附图说明Description of drawings
图1为本发明提供的面向多类型供用能系统调控运行知识图谱构建流程图。Fig. 1 is a flow chart of the construction of knowledge map for regulation and operation of multi-type energy supply and utilization systems provided by the present invention.
图2为本发明提供的知识图谱模式层结构图；Fig. 2 is the structural diagram of the knowledge map mode layer provided by the present invention;
图3为本发明实施例提供的知识图谱数据层结构图；FIG. 3 is a structural diagram of a knowledge map data layer provided by an embodiment of the present invention;
图4为本发明实施例提供的知识图谱应用流程结构图；FIG. 4 is a structural diagram of a knowledge graph application process provided by an embodiment of the present invention;
图5为本发明实施例提供的实施例知识图谱展示图。Fig. 5 is a display diagram of an example knowledge map provided by an example of the present invention.
具体实施方式Detailed ways
下面结合附图和实施例，对本发明的具体实施方式作进一步详细描述。以下实施例用于说明本发明，但不用来限制本发明的范围。The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
如图1所示，本实施例的面向多类型供用能系统调控运行知识图谱构建方法如下所述。As shown in FIG. 1 , the construction method of the multi-type energy supply and utilization system regulation and operation knowledge graph of this embodiment is as follows.
步骤1：确定多类型供用能系统调控运行策略的运行技术指标，并依次细化，采用自顶向下的方法构建知识图谱模式层。知识图谱模式层存储概念、规则、公理和约束条件，模式层中的实体一般为电力系统中经过提炼的抽象名词，也称本体，多类型供用能系统调控运行策略下详细业务的逻辑关系复杂，实体较多，需要先抽象出本体之间的关系，然后通过实体的学习构建出最后的业务。图2为知识图谱模式层基本结构，依据多类型供用能系统配电网实际运行状态和调控需求，确定运行技术指标为资源类型、组合方式、用户组成、运行指标等并向下细化以完成知识图谱模式层的构建。资源类型细化为光伏发电、风力发电、生物质能等。组合方式细化为电网+光伏发电、电网+热电联产发电+并网上网等。用户组成细化为乡村居民、加工车间、养殖场等。运行指标细化为经济运行、稳定运行。Step 1: Determine the operation technical indicators of the multi-type energy supply and use system regulation and operation strategy, and refine them in turn, and use the top-down method to construct the knowledge map model layer. The schema layer of the knowledge map stores concepts, rules, axioms, and constraints. Entities in the schema layer are generally refined abstract nouns in the power system, also known as ontology. The logical relationship of detailed business under the regulation and operation strategy of multi-type energy supply and utilization systems is complex. There are many entities, and it is necessary to abstract the relationship between the ontology first, and then construct the final business through entity learning. Figure 2 shows the basic structure of the knowledge map model layer. According to the actual operation status and regulation requirements of the multi-type energy supply and utilization system distribution network, the operation technical indicators are determined as resource type, combination mode, user composition, operation indicators, etc., and are refined downwards to complete The construction of knowledge map pattern layer. Resource types are subdivided into photovoltaic power generation, wind power generation, biomass energy, etc. The combination mode is refined into grid + photovoltaic power generation, grid + combined heat and power generation + grid connection, etc. The composition of users is subdivided into rural residents, processing workshops, farms, etc. The operation indicators are refined into economic operation and stable operation.
步骤2：对调控运行策略数据进行数据预处理。调控运行策略数据包括运行要求、调控预案、运行报告，为加快神经网络模型训练速度，提高神经网络模型精确度，需要对调控运行策略数据进行清洗、删除与补充，并将调控运行策略数据按照特征指标划分为训练样本与测试样本。对调控运行策略数据进行分词，以得到调控词库，并根据调控词库，构建空间向量。首先确定查询文本预处理之后所含的每一个文档，并根据文档所属领域词库，对每一个文档进行分词，将每一个文档中的每个所属的备选关键词代入所得到的分析模型中，得到每一个文档中各备选关键词对应形成的若干维度的词向量，并作为空间向量输出。Step 2: Carry out data preprocessing on the control and operation strategy data. The regulation and operation strategy data includes operation requirements, regulation and control plans, and operation reports. In order to speed up the training speed of the neural network model and improve the accuracy of the neural network model, it is necessary to clean, delete and supplement the regulation and operation strategy data, and divide the regulation and operation strategy data according to the characteristics The indicators are divided into training samples and testing samples. Segment the control and operation strategy data to obtain the control lexicon, and construct the space vector according to the control lexicon. First, determine each document included in the query text after preprocessing, and perform word segmentation on each document according to the thesaurus of the field to which the document belongs, and substitute each candidate keyword in each document into the obtained analysis model , to obtain word vectors of several dimensions corresponding to each candidate keyword in each document, and output them as space vectors.
步骤3：构建不同的神经网络模型以实现调控运行策略数据的知识抽取并得到包含“实体-属性-实体”的三元组数据。Step 3: Construct different neural network models to realize the knowledge extraction of control operation strategy data and obtain the triple data containing "entity-attribute-entity".
步骤3.1：基于字向量使用TextCNN模型对调控运行策略数据进行文本分类。文本分类是将调控运行策略数据文本按预先定义的类别进行自动分类标记。TextCNN模型将调控运行策略数据文本中的电网运行状态进行划分。Step 3.1: Use the TextCNN model to classify the text of the control operation strategy data based on the word vector. Text classification is to automatically classify and mark the control operation policy data text according to the predefined categories. The TextCNN model divides the power grid operation status in the control operation strategy data text.
步骤3.2：使用LR-CNN模型实现对调控运行策略数据的命名实体识别。命名实体识别是对调控运行策略数据文本中具有特定含义的实体进行边界确定和类别识别，根据乡村配电网中包含的实体属性与种类，使用LR-CNN模型将其中实体划分为设备名称、资源名称、投资金额、回收周期、运行地点等。其中，本发明使用的LR-CNN模型用堆叠的窗口大小为2的CNN模型对句子的字符、候选词特征进行提取，并使用注意力机制模块合并字符和候选词信息。同时引入 Rethinking 机制，向每个 CNN 层添加反馈层，使用高层的字词信息来调整低层的注意力机制模块的权重，降低错误候选词的权重解决候选词冲突问题。Step 3.2: Use the LR-CNN model to realize the named entity recognition of the regulatory operation policy data. Named entity recognition is to determine the boundaries and identify the entities with specific meanings in the data text of the regulation and operation strategy. According to the attributes and types of entities contained in the rural power distribution network, the LR-CNN model is used to divide the entities into equipment names, resource Name, investment amount, payback period, operating location, etc. Wherein, the LR-CNN model used in the present invention uses a CNN model with a stacked window size of 2 to extract the character and candidate word features of the sentence, and uses the attention mechanism module to merge the character and candidate word information. At the same time, the Rethinking mechanism is introduced, adding a feedback layer to each CNN layer, using high-level word information to adjust the weight of the low-level attention mechanism module, and reducing the weight of wrong candidate words to solve the problem of candidate word conflicts.
步骤3.3：运用BERT-BiLSTM-CRF模型对调控运行策略数据中的实体进行实体间关系抽取。实体间关系抽取是在命名实体识别的基础上判断实体间是否存在预定义的关系，从而构成一系列三元组知识。在本发明中，通过配电网运行规范与状态对实体间关系进行预定义，利用抽取后形成的关系构建知识图谱中的一系列三元组，同时本发明使用预训练过的 BERT 模型，在后面连接 BiLSTM-CRF 模型，同时将 BERT 层的输出和 BiLSTM 层的输出进行特征串联，并对其进行微调训练，进一步提高关系抽取的准确率。Step 3.3: Use the BERT-BiLSTM-CRF model to extract the relationship between entities in the control operation strategy data. Inter-entity relationship extraction is based on named entity recognition to determine whether there is a predefined relationship between entities, thereby forming a series of triples of knowledge. In the present invention, the relationship between entities is predefined through the operation specification and state of the distribution network, and a series of triples in the knowledge graph are constructed using the relationship formed after extraction. At the same time, the present invention uses the pre-trained BERT model to The BiLSTM-CRF model is connected later, and at the same time, the output of the BERT layer and the output of the BiLSTM layer are concatenated, and fine-tuned training is performed to further improve the accuracy of relationship extraction.
步骤4：将三元组数据加入知识图谱中并采用自底向上的方法构建知识图谱数据层；图3为知识图谱数据层结构，主要包括设备状态、设备出力、运行时间、运行成本、地点等。在完成调控运行策略数据的文本分类、命名实体识别、实体间关系抽取之后，将得到的包含“实体-属性-实体”的三元组加入知识图谱数据层对应位置。Step 4: Add the triplet data into the knowledge map and use the bottom-up method to construct the knowledge map data layer; Figure 3 shows the structure of the knowledge map data layer, which mainly includes equipment status, equipment output, running time, running cost, location, etc. . After completing the text classification, named entity recognition, and entity-to-entity relationship extraction of the regulatory operation strategy data, the obtained triples containing "entity-attribute-entity" are added to the corresponding position of the knowledge graph data layer.
步骤5：将知识图谱数据层数据进行归类并与知识图谱模式层结合形成策略集，构建出完整的知识图谱并存储。Step 5: Classify the knowledge map data layer data and combine it with the knowledge map pattern layer to form a policy set, construct a complete knowledge map and store it.
步骤5.1：基于多类型供用能系统配电网资源种类及特性确定典型场景。基于用户冷热电需求特征数据、可再生能源资源评估数据，将适宜的用户进行聚合形成不同类型的能源合作社，并基于多类型供用能系统配电网运行特点将部分能源合作社确定为典型场景，例如乡村生活区能源合作社、生态旅游能源合作社、生态大棚种植能源合作社等。Step 5.1: Determine typical scenarios based on the types and characteristics of distribution network resources in multi-type energy supply and consumption systems. Based on the characteristic data of users' demand for cooling, heating and power, and the evaluation data of renewable energy resources, suitable users are aggregated to form different types of energy cooperatives, and some energy cooperatives are determined as typical scenarios based on the operation characteristics of distribution networks of multi-type energy supply and consumption systems. For example, rural living area energy cooperatives, ecotourism energy cooperatives, ecological greenhouse planting energy cooperatives, etc.
步骤5.2：将知识图谱数据层中的调控要求整理为策略集。通过对知识图谱数据层中的调控要求进行评估与判断，例如用户组合方式、资源运行季节、资源互补关系、设备运行状态等，将调控要求加入对应典型场景下形成策略集。Step 5.2: Organize the regulatory requirements in the knowledge map data layer into a policy set. By evaluating and judging the regulatory requirements in the knowledge map data layer, such as user combination mode, resource operating season, resource complementary relationship, equipment operating status, etc., the regulatory requirements are added to the corresponding typical scenarios to form a policy set.
步骤5.3：将知识图谱数据层中的具体数据归纳为策略并放入策略集。过对知识图谱数据层中的具体数据，例如冷、热、电设计负荷、供能系统装机方案、全年能耗量、总投资等数据进行归纳提取并将其加入到对应策略集中，形成策略集中可辅助电网工作人员进行电力调度、参考决策的详细策略信息。Step 5.3: Summarize the specific data in the knowledge graph data layer into strategies and put them into the strategy set. By summarizing and extracting specific data in the data layer of the knowledge map, such as cold, heat, and electrical design loads, energy supply system installation plans, annual energy consumption, and total investment, and adding them to the corresponding strategy set, a strategy is formed Concentrate detailed policy information that can assist power grid staff in power dispatching and reference decision-making.
步骤5.4：将知识图谱数据层与知识图谱模式层结合形成完整的知识图谱。Step 5.4: Combine the knowledge graph data layer with the knowledge graph schema layer to form a complete knowledge graph.
确定的知识图谱模式层中的主要运行指标，将策略与策略集及对应的知识图谱数据层数据嵌入在对应的知识图谱模式层的结构中，形成完整的知识图谱。Determine the main operating indicators in the knowledge map mode layer, and embed the strategy and strategy set and the corresponding knowledge map data layer data in the structure of the corresponding knowledge map mode layer to form a complete knowledge map.
步骤5.5：存储知识图谱。将构建的完整的知识图谱存储在Neo4j图数据库中,Neo4j是基于Java语言编写的图形数据库，它采用节点和关系的形式存储信息，并在此基础上提供界面友好的可视化演示。Step 5.5: Store the knowledge graph. Store the completed knowledge map in the Neo4j graph database. Neo4j is a graph database written based on the Java language. It stores information in the form of nodes and relationships, and provides a friendly visual presentation on this basis.
一种面向多类型供用能系统调控运行知识图谱的应用，在季节性运行触发条件下，找到知识图谱中对应内容并进行细化分解，通过逐级构建，结合已有的数据与不同的条件，继续向下生成不同典型场景并细化为对应策略集，并将知识图谱数据层数据加入对应方案集的方案中输出，生成完整的调控运行策略辅助工作人员进行决策。参照图4，具体步骤如下：An application oriented to multi-type energy supply and utilization system regulation and operation knowledge map. Under the trigger condition of seasonal operation, find the corresponding content in the knowledge map and carry out detailed decomposition. Through step-by-step construction, combined with existing data and different conditions, Continue to generate different typical scenarios downwards and refine them into corresponding policy sets, and add the knowledge map data layer data to the corresponding scheme set for output, and generate a complete regulation and operation strategy to assist staff in making decisions. Referring to Figure 4, the specific steps are as follows:
步骤一：接收季节性运行触发条件并在知识图谱中定位。在多类型供用能系统配电网运行的过程中，接收到知识图谱的季节性运行触发条件指令或要求，例如夏季运行状态、冬季运行状态、节假日运行状态、临时规定等要求，电网工作人员在接到触发条件后，在知识图谱存储的Neo4j图数据库中进行关键字或词检索，找到对应知识图谱所在位置，完成对季节新运行触发指令下知识图谱的定位。Step 1: Receive seasonal operation trigger conditions and locate them in the knowledge graph. During the operation of the multi-type energy supply and consumption system distribution network, the seasonal operation trigger condition instructions or requirements of the knowledge map are received, such as summer operation status, winter operation status, holiday operation status, temporary regulations, etc. After receiving the trigger conditions, search for keywords or words in the Neo4j graph database stored in the knowledge graph, find the location of the corresponding knowledge graph, and complete the positioning of the knowledge graph under the new seasonal operation trigger command.
步骤二：选择配电网运行指标。在完成定位以后，结合配电网运行状态与调控需求，继续向下选择配电网运行指标，例如经济运行、资源平抑、线路重过载、电压越限、稳定运行等指标。Step 2: Select the distribution network operation index. After the positioning is completed, combined with the distribution network operation status and regulation requirements, continue to select distribution network operation indicators downward, such as economic operation, resource stabilization, line heavy overload, voltage limit, stable operation and other indicators.
步骤三：生成各运行指标下对应的典型场景。确定电网运行技术指标后，在相应指标下继续细化，生成对应的典型场景，如生态旅游合作社，养殖场能源合作社，农产品加工合作社等。Step 3: Generate typical scenarios corresponding to each operating indicator. After determining the technical indicators of power grid operation, continue to refine under the corresponding indicators to generate corresponding typical scenarios, such as eco-tourism cooperatives, farm energy cooperatives, agricultural product processing cooperatives, etc.
步骤四：得到对应的策略集与调控运行策略。在生成的对应的典型场景下选择其中的典型用户，例如乡村居民、学校、零售商店、行政服务、铁塔、网络服务、生活动力设施、生活生产服务、卫生所、加工车间等，接下来对典型用户供用能特性及资源数据进行细分，得到包含用户组合方式、资源运行季节、资源互补关系、设备运行状态等信息的策略集，接着对知识图谱数据层中具体数据例如冷、热、电设计负荷、供能系统装机方案、全年能耗量、总投资等数据进行归纳提取并将其加入到对应策略集中，形成策略集中可辅助电网工作人员进行电力调度、参考决策的详细策略信息。Step 4: Obtain the corresponding strategy set and regulation operation strategy. Select typical users in the generated corresponding typical scenarios, such as rural residents, schools, retail stores, administrative services, iron towers, network services, life power facilities, life production services, health centers, processing workshops, etc. Subdividing user energy supply and consumption characteristics and resource data to obtain a policy set including information such as user combination mode, resource operating season, resource complementary relationship, and equipment operating status, and then specific data in the knowledge graph data layer such as cold, heat, and electricity design Load, energy supply system installed capacity plan, annual energy consumption, total investment and other data are summarized and extracted and added to the corresponding strategy set to form detailed strategy information that can assist power grid staff in power dispatching and reference decision-making.
在知识图谱应用过程中，逐步添加对应信息完成对知识图谱模式层与知识图谱数据层的更新；电力系统中的知识在不断增加和更新，知识图谱在建成后还需要动态构建和迭代更新，不断增加新的知识、删除旧的知识并相应调整知识图谱的结构。知识图谱的更新包括知识图谱数据层的更新和知识图谱模式层的更新。相对而言，知识图谱数据层更新对知识图谱的整体架构影响较小，而知识图谱模式层更新的影响较大；因此，知识图谱数据层往往可以采取自动化的更新方式，而知识图谱模式层更新则往往需要人工确认和审核。In the application process of the knowledge map, corresponding information is gradually added to complete the update of the knowledge map model layer and the knowledge map data layer; the knowledge in the power system is constantly increasing and updating, and the knowledge map needs to be dynamically constructed and iteratively updated after it is completed. Add new knowledge, delete old knowledge, and adjust the structure of the knowledge graph accordingly. The update of the knowledge map includes the update of the data layer of the knowledge map and the update of the pattern layer of the knowledge map. Relatively speaking, the update of the knowledge map data layer has little impact on the overall structure of the knowledge map, while the update of the knowledge map model layer has a greater impact; therefore, the knowledge map data layer can often be updated in an automated way, while the update of the knowledge map model layer Manual confirmation and review is often required.
对于知识图谱模式层的更新，采用专家评估的方式，根据配电网一段时间内的运行状态与未来配电网运行要求来确定模式层中需要的更新内容，同时采用增量更新的方法更新模式层，增量更新是以新增数据作为输入对知识图谱进行更新，其资源消耗较小。For the update of the knowledge map model layer, expert evaluation is adopted to determine the update content required in the model layer according to the operation status of the distribution network for a period of time and the future operation requirements of the distribution network, and the incremental update method is used to update the model Layer, incremental update is to update the knowledge map with new data as input, and its resource consumption is small.
知识图谱数据层同样采用增量更新的方法，知识图谱数据层中调控运行策略文档在每一年都会进行更新，在同一年中也会季度性更新，同时可能还有一些节假日、临时规定。在每一版本的调控运行策略文档中，存在多条规定，对于每一条规定而言，又存在多层运行方式定义，最底层的方式使用调控运行策略和策略集表来表达调控运行策略的运行范围。The knowledge map data layer also adopts the method of incremental update. The regulation and operation policy documents in the knowledge map data layer will be updated every year, and will also be updated quarterly in the same year. At the same time, there may be some holidays and temporary regulations. In each version of the regulatory operation strategy document, there are multiple regulations. For each regulation, there are multi-layered operation mode definitions. The bottom-level method uses the regulatory operation strategy and the policy set table to express the operation of the regulatory operation strategy. scope.
综上，本实施例，针对多类型供用能系统配电网运行自动化程度不高，运行调控信息冗杂的问题，利用知识图谱技术对调控运行策略信息进行知识抽取、表示和管理，并用于辅助调度人员进行电网调控，可以提高决策效率与决策的生成速度，同时可以更直观的展示不同策略之间的关联关系，提升了电网应急处理能力与调度智能化水平。以电网调控运行策略预案文本为研究对象，提出了一种自顶向下和自底向上相结合的多类型供用能系统调控运行知识图谱构建方法，并解决了其中涉及的电力领域知识抽取问题。首先，自顶向下定义知识图谱的知识组织架构、概念类型、概念间关系，形成知识图谱的模式层；之后，针对电网故障处置预案文本的特性，综合使用多种深度学习模型进行知识抽取，自底向上构建知识图谱的数据层：为避免分词错误，使用基于字向量的TextCNN 模型对预案文本进行分类；为解决候选词冲突问题，使用 LR-CNN 模型对电力领域的命名实体进行识别； 在命名实体识别的基础上，使用BERT- BiLSTM-CRF模型对实体间关系进行抽取。之后，通过实验验证了上述知识抽取方法的有效性。最后，对构建的知识图谱进行了可视化并对其在调控策略生成的应用进行了分析。To sum up, in this embodiment, aiming at the problems of low degree of automation in the distribution network operation of multi-type energy supply and consumption systems and redundant operation control information, the knowledge map technology is used to extract, represent and manage knowledge control operation strategy information, and use it to assist dispatching Power grid regulation by personnel can improve decision-making efficiency and decision-making speed. At the same time, it can more intuitively display the correlation between different strategies, and improve the power grid emergency handling capability and intelligent dispatching level. Taking the power grid regulation and operation strategy pre-plan text as the research object, a top-down and bottom-up combination of multi-type energy supply and consumption system regulation and operation knowledge map construction method is proposed, and the problem of knowledge extraction in the power field involved is solved. First, define the knowledge organization structure, concept types, and inter-concept relationships of the knowledge graph from top to bottom to form the model layer of the knowledge graph; then, according to the characteristics of the power grid fault handling plan text, a variety of deep learning models are used for knowledge extraction. Build the data layer of the knowledge map from the bottom up: in order to avoid word segmentation errors, use the TextCNN model based on word vectors to classify the plan text; in order to solve the problem of candidate word conflicts, use the LR-CNN model to identify named entities in the power field; Based on named entity recognition, the BERT-BiLSTM-CRF model is used to extract the relationship between entities. Afterwards, the effectiveness of the above knowledge extraction method is verified through experiments. Finally, the constructed knowledge graph is visualized and its application in regulatory policy generation is analyzed.
以多类型供用能系统配电网中某乡村典型场景中生态林场下应用知识图谱为例，图5为本发明实施例提供的在生态林场下应用知识图谱的示意图。Taking the knowledge map applied in an ecological forest farm in a typical rural scene in a multi-type energy supply and utilization system distribution network as an example, FIG. 5 is a schematic diagram of the knowledge map applied in an ecological forest farm provided by an embodiment of the present invention.
表1 多类型供用能系统中生某乡村生态林场经济运行场景下调控运行策略具体实施方案表Table 1 The specific implementation plan of the regulation and operation strategy under the economic operation scenario of a rural ecological forest farm in a multi-type energy supply and use system
首先确定该典型场景即生态林场内的主要能源合作社类型，即养殖场能源合作社，然后将经济运行确定为在该典型场景下配电网运行的主要技术指标，通过对在该典型场景下的调控运行策略预案数据进行文本分类，实体识别，实体间关系抽取后形成相应的三元组加入到知识图谱中形成该典型场景下的知识图谱数据层，其中，主要包括合作社内总负荷和养殖场供能资源，合作社内总负荷又将细化为设备类型与负荷类别，设备类型中包括供暖设备和供冷设备，继续分别将供暖设备细分为电灯板和暖灯，将供冷设备细分为冷风机和负压风机，负荷类别继续细化得到养殖栏舍供暖、养殖栏舍供冷、办公楼供冷、办公楼供暖，养殖场供能资源被细化为沼气、地热、电力，接着将各类别下的具体数据整理并加入到知识图谱对应位置，这部分将作为知识图谱的输入部分，通过可视化展示，供电网工作人员对此养殖场能源合作社场景下的供用能信息作参考与了解。First, determine the typical scenario, that is, the main type of energy cooperatives in the ecological forest farm, that is, farm energy cooperatives, and then determine the economic operation as the main technical index for the operation of the distribution network in this typical scenario. Run the strategy plan data for text classification, entity recognition, and extract the relationship between entities to form corresponding triples and add them to the knowledge map to form the knowledge map data layer in this typical scenario, which mainly includes the total load in the cooperative and the supply of farms. Energy resources, the total load in the cooperative will be subdivided into equipment types and load categories. Air coolers and negative pressure fans, the load category continues to be refined to obtain heating for breeding pens, cooling for breeding pens, cooling for office buildings, and heating for office buildings. The energy supply resources for farms are refined into biogas, geothermal, and electricity. The specific data under each category is sorted and added to the corresponding position of the knowledge map. This part will be used as the input part of the knowledge map. Through visual display, the power supply network staff can refer to and understand the energy supply and consumption information in the farm energy cooperative scenario.
接下来将调控运行策略预案数据中涉及到调控运行策略的内容进行归纳整理并将各调控运行策略整合为策略集，在生态林场经济运行条件下生成的策略集为：沼气系统+光伏系统+栏舍热泵加热制冷，策略集中的具体策略为：沼气系统增加发电余热回收装置，提高冬季沼气产量；光伏系统采用全面积铺装，自用为主，余电上网；栏舍热泵加热制冷系统为办公室采用空气源热泵。Next, the content related to the regulation and operation strategy in the regulation and operation strategy pre-plan data is summarized, and the regulation and operation strategies are integrated into a strategy set. The strategy set generated under the economic operation conditions of the ecological forest farm is: biogas system + photovoltaic system + column The heat pump heating and cooling of the house, the specific strategy of strategy concentration is as follows: the biogas system increases the power generation waste heat recovery device to increase the biogas production in winter; the photovoltaic system adopts full-area pavement, mainly for self-use, and the surplus electricity is connected to the grid; the barn heat pump heating and cooling system is adopted for the office Air source heat pump.
最后将调控运行策略中的具体信息归纳总结为包含各资源出力与装机方案及总投资和回收周期的具体实施方案输出，此场景下生成的具体实施方案如表1所示，由此构成完整的针对生态林场下的知识图谱应用过程。Finally, the specific information in the control and operation strategy is summarized as the output of the specific implementation plan including the output of each resource, the installed capacity plan, the total investment and the recovery cycle. The specific implementation plan generated in this scenario is shown in Table 1, thus forming a complete Aiming at the application process of knowledge map under ecological forest farm.
本发明还提供一种面向多类型供用能系统调控运行知识图谱构建方法的装置，包括知识图谱模式层模块、数据预处理模块、知识抽取模块、知识图谱数据层模块、策略集模块和存储模块；The present invention also provides a device oriented to the construction method of multi-type energy supply and utilization system control and operation knowledge graph, including a knowledge graph model layer module, a data preprocessing module, a knowledge extraction module, a knowledge graph data layer module, a strategy set module and a storage module;
所述知识图谱模式层模块用于确定多类型供用能系统调控运行策略的运行技术指标，并依次细化形成知识图谱模式层；The knowledge map model layer module is used to determine the operation technical indicators of the multi-type energy supply and utilization system regulation and control operation strategy, and successively refine the knowledge map model layer;
所述数据预处理模块用于对调控运行策略数据进行数据预处理；The data preprocessing module is used to perform data preprocessing on the regulation and control operation strategy data;
所述知识抽取模块用于构建不同的神经网络模型以实现调控运行策略数据的知识抽取并得到包含实体-属性-实体的三元组数据；The knowledge extraction module is used to construct different neural network models to realize the knowledge extraction of control operation strategy data and obtain triple data comprising entity-attribute-entity;
所述知识图谱数据层模块将三元组数据加入知识图谱中并采用自底向上的方法构建知识图谱数据层；The knowledge graph data layer module adds triplet data into the knowledge graph and adopts a bottom-up method to construct the knowledge graph data layer;
所述策略集模块将知识图谱数据层数据进行归类并与知识图谱模式层结合形成策略集；所述存储模块用于存储知识图谱。The strategy set module classifies the knowledge graph data layer data and combines it with the knowledge graph pattern layer to form a strategy set; the storage module is used to store the knowledge graph.
本发明提供了一种计算机存储介质，所述计算机存储介质上存储有计算机程序指令；所述计算机程序指令被处理器执行时实现面向多类型供用能系统调控运行知识图谱构建方法。The present invention provides a computer storage medium, on which computer program instructions are stored; when the computer program instructions are executed by a processor, a method for constructing a knowledge map oriented to multi-type energy supply and utilization system regulation and operation is implemented.
最后应说明的是：以上各实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述各实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明权利要求所限定的范围。Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: It is still possible to modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope defined by the claims of the present invention .
Claims (8)
Translated fromChinesePriority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211272027.7ACN115344717B (en) | 2022-10-18 | 2022-10-18 | Method and device for constructing knowledge map for multi-type energy supply and utilization system regulation and operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211272027.7ACN115344717B (en) | 2022-10-18 | 2022-10-18 | Method and device for constructing knowledge map for multi-type energy supply and utilization system regulation and operation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115344717A CN115344717A (en) | 2022-11-15 |
| CN115344717Btrue CN115344717B (en) | 2023-02-17 |
Family
ID=83957099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211272027.7AActiveCN115344717B (en) | 2022-10-18 | 2022-10-18 | Method and device for constructing knowledge map for multi-type energy supply and utilization system regulation and operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115344717B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117150040A (en)* | 2023-08-30 | 2023-12-01 | 中国电力科学研究院有限公司 | A method and system for constructing security control strategy knowledge graph for AC and DC power grids |
| CN118446835B (en)* | 2024-04-02 | 2024-12-13 | 中国长江电力股份有限公司 | Intelligent operation method and system for step power station |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112148893A (en)* | 2020-09-25 | 2020-12-29 | 南方电网数字电网研究院有限公司 | Energy analysis knowledge graph construction method and energy analysis visualization method |
| CN112612902A (en)* | 2020-12-23 | 2021-04-06 | 国网浙江省电力有限公司电力科学研究院 | Knowledge graph construction method and device for power grid main device |
| CN112905804A (en)* | 2021-02-22 | 2021-06-04 | 国网电力科学研究院有限公司 | Dynamic updating method and device for power grid dispatching knowledge graph |
| CN114077674A (en)* | 2021-10-31 | 2022-02-22 | 国电南瑞科技股份有限公司 | Power grid dispatching knowledge graph data optimization method and system |
| CN115033705A (en)* | 2022-04-21 | 2022-09-09 | 国网江苏省电力有限公司南京供电分公司 | Design method and system of knowledge map for power grid regulation risk early warning information |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200226133A1 (en)* | 2016-10-18 | 2020-07-16 | Hithink Financial Services Inc. | Knowledge map building system and method |
| US11327989B2 (en)* | 2017-08-02 | 2022-05-10 | Accenture Global Solutions Limited | Multi-dimensional industrial knowledge graph |
| CN111930784B (en)* | 2020-07-23 | 2022-08-09 | 南京南瑞信息通信科技有限公司 | Power grid knowledge graph construction method and system |
| CN112242187B (en)* | 2020-10-26 | 2023-06-27 | 平安科技（深圳）有限公司 | Medical scheme recommendation system and method based on knowledge graph characterization learning |
| CN115170344A (en)* | 2022-05-30 | 2022-10-11 | 国网江苏省电力有限公司南京供电分公司 | Intelligent processing method and device, medium and equipment for operation events of regulation and control system |
- 2022
- 2022-10-18CNCN202211272027.7Apatent/CN115344717B/enactiveActive
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112148893A (en)* | 2020-09-25 | 2020-12-29 | 南方电网数字电网研究院有限公司 | Energy analysis knowledge graph construction method and energy analysis visualization method |
| CN112612902A (en)* | 2020-12-23 | 2021-04-06 | 国网浙江省电力有限公司电力科学研究院 | Knowledge graph construction method and device for power grid main device |
| CN112905804A (en)* | 2021-02-22 | 2021-06-04 | 国网电力科学研究院有限公司 | Dynamic updating method and device for power grid dispatching knowledge graph |
| CN114077674A (en)* | 2021-10-31 | 2022-02-22 | 国电南瑞科技股份有限公司 | Power grid dispatching knowledge graph data optimization method and system |
| CN115033705A (en)* | 2022-04-21 | 2022-09-09 | 国网江苏省电力有限公司南京供电分公司 | Design method and system of knowledge map for power grid regulation risk early warning information |
Non-Patent Citations (6)
| Title |
|---|
| Construction technology of large scale knowledge maps in customer service field of Power Grid;Shuang Shuang Guo等;《2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)》;20210510;全文* |
| 基于知识图谱技术的风电数据管理与应用研究;吕梦平;《CNKI优秀硕士学位论文全文库》;20220515;全文* |
| 基于知识图谱的电力知识平台关键技术研究;俞阳;《CNKI优秀硕士学位论文全文库》;20190515;全文* |
| 知识图谱在智能电网的应用;王琼等;《电子元器件与信息技术》;20200120(第01期);全文* |
| 调度自动化系统知识图谱的构建与应用;李新鹏等;《中国电力》;20181208(第02期);全文* |
| 面向智能调控领域的知识图谱构建与应用;余建明等;《电力系统保护与控制》;20200201(第03期);全文* |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115344717A (en) | 2022-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115344717B (en) | Method and device for constructing knowledge map for multi-type energy supply and utilization system regulation and operation | |
| CN113377966B (en) | Water conservancy project scheduling regulation reasoning method based on knowledge graph | |
| CN111522950B (en) | A Rapid Identification System for Unstructured Massive Text Sensitive Data | |
| CN112685570B (en) | Multi-label graph-based power grid network frame knowledge graph construction method | |
| CN117934209B (en) | Regional power system carbon emission big data analysis method based on knowledge graph | |
| CN115757810B (en) | A knowledge graph standard ontology construction method | |
| CN116127084A (en) | Intelligent retrieval system and method for microgrid dispatching strategy based on knowledge graph | |
| CN115705364A (en) | Power grid monitoring knowledge graph construction method based on equipment operation characteristics | |
| CN115033705A (en) | Design method and system of knowledge map for power grid regulation risk early warning information | |
| CN114491168A (en) | Method and system for regulating and controlling cloud sample data sharing, computer equipment and storage medium | |
| CN116775895A (en) | A wind turbine operation and maintenance management method based on knowledge graph | |
| CN117194671A (en) | Construction method and system of equipment maintenance disassembly and assembly process knowledge graph | |
| CN116340534A (en) | A knowledge map construction method and system for identifying new energy abnormal data | |
| CN112668836B (en) | Risk spectrum-oriented associated risk evidence efficient mining and monitoring method and apparatus | |
| Zhang et al. | Forecasting Chinese carbon emissions from fossil energy based on the fractional order cumulative multivariate Grey model | |
| CN115510961A (en) | Community comprehensive energy system operation safety assessment method based on active learning | |
| Zhenyu et al. | Research on Construction and Application of Regulation of the Multiple Energy Systems Based on Knowledge Graph | |
| Li et al. | Knowledge Base Question Answering for Intelligent Maintenance of Power Plants | |
| Gu et al. | Research on the Knowledge Graph Construction Technology and its Power Grid Applications | |
| Xiong et al. | Experimental Evaluation of Intelligent e-Government System Based on Text Mining | |
| Yan et al. | Construction of Dynamic Knowledge Graph for Grid day-ahead Scheduling Optimization Decision | |
| CN111651496A (en) | A system and method for repeated power failure analysis and decision-making based on artificial intelligence | |
| CN120181243B (en) | Knowledge graph reasoning method and device based on text mapping and vector library retrieval | |
| CN113191735B (en) | Case-driven-based power grid asset life-span management method and system | |
| Chen et al. | Construction of Knowledge Map of Power Marketing User Address Data Based on Information Extraction Model |
Legal Events
| 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 |