Research On Semantic Parsing And Assisted Decision Making Scheme Of Power Grid Dispatching And Monitoring Signal

The power dispatching system contains a large number of text form of record information such as dispatching operation instructions,monitoring exception record,dispatching and monitoring signal,etc.These texts contain rich information,which can reflect the operation state of power grid and the reliability level of equipment,and provide support for intelligent dispatching of power grid.Dispatching and monitoring signal is one of the important text of power system.In this paper,the text of dispatching and monitoring signal in power grid is taken as the research object.Combining the existing semantic analysis technology and power system expertise,a semantic parsing method is proposed.Combined with data mining technology,an assisted decision making scheme is proposed.Firstly,according to the relevant standards provided by IEC and the historical data provided by a power enterprise in East China,a corpus including sample database and physical environment database is established.The corpus combines the characteristics of the text itself,and is the basis of semantic parsing of the text.A dictionary set including professional dictionary and standard dictionary is established,which provides support for Chinese word segmentation and chunk analysis of monitoring signal text.Based on corpus and dictionaries,a semantic parsing method is proposed.Firstly,the whole framework of semantic parsing is introduced,and the research objectives are divided into two main function points:Chinese word segmentation and semantic parsing.According to the proposed function points,the method is divided into dictionary loading module,formal module,word segmentation module,abstract module,combination module and semantic parsing module.Combined with the actual data,the results of each module are introduced.The experimental results show that this method has good segmentation effect and entity recognition effect,and finally obtains the XML file which can be parsed by computer,which lays a foundation for the intelligent dispatching of power grid.On the basis of semantic analysis,an assisted decision making scheme based on Apriori algorithm is proposed.The Apriori algorithm is used to mine the potential relationship between the monitoring signals and the master-slave law.Then combining the theoretical knowledge and using k-means clustering algorithm to define the signal type of monitoring signal and set the related threshold value.Combining with the historical data,the expert database template and the template prefix tree are established,and finally,the above results are used to make the correct auxiliary decision for real-time events.The semantic parsing method and the assisted decision making scheme proposed in this paper are verified in real data,and the experimental results are good,which verify the effectiveness and feasibility of the proposed method.