Research On Method Of Multi-Source Heterogeneous Data Fusion For Electric Power Equipment

With the development of state monitoring technology and the interaction of various automation information application systems in the power grid,the power equipment data presents features such as large data size,fast update speed,multi-source heterogeneity,and low value density,which is in full compliance with all big data.Characteristics,typical big data.Under the new situation of explosive growth of power equipment big data,traditional data processing technology encounters bottlenecks and cannot meet the demand for rapid acquisition of knowledge and information from massive data.Power equipment data fusion is the information and intelligence of the power industry.The inevitable requirements of development.With the development of big data fusion technology,there are more and more methods for multi-source heterogeneous big data fusion.Research on data fusion of multi-source heterogeneous big data in power equipment becomes a research hotspot for the further development of smart grid,and for smart grids Stable,safe and reliable operation is of great significance.This article studies the key technologies of power equipment data fusion,analyzes the problems of power equipment big data fusion and the basic principles of different data fusion technologies.Afterwards,the model heterogeneity of public information model and power equipment state information is studied.Based on the data fusion processing technology,the similarities and differences between the public information model and the power equipment state information model are analyzed,and a heterogeneous model alignment method is designed.Based on this,a unified information model for power equipment states was built,and resource sharing among heterogeneous systems was realized.Aiming at the problems of data noise interference and large amount of redundant data in the state monitoring data of power equipments,the Kalman filter algorithm of the classical data fusion method is improved,and an event-driven distributed Kalman filter algorithm is proposed.The algorithm adds event decision strategy during the data processing and transmission between sensor nodes,and controls the number of communication between nodes in an event-driven manner.At the same time,it reduces the redundant data and noise-interfered data through the processing of the algorithm,while ensuring the data quality.,improve the integration efficiency.Finally,by comparing with the conventional distributed Kalman filtering algorithm,the effectiveness of the algorithm is verified by experimental simulation.