Research And Application Of Multi-source Information Fusion Technology Based On Big Data In Power System

With the continuous upgrading and upgrading of the intelligent power system,the number and type of big data have begun to show an explosive growth trend,and the development of power grid in China has e ntered the era of big data.How to collect useful information quickly in a chaotic data pile is a huge challenge for power system construction.Multi-source information fusion plays an important role in the research of big data.It can integrate the data and information which belongs to the research object from many aspects.It can realize the disorganized grid data and get more detailed and comprehensive analysis of the system.And relying on the mature development of storage and processing technology of big data,it provides new opportunities for the research of multi-source information fusion of big data.Based on the classification of source network load storage,the distribution of big data in power system and its "4V" characteristics are briefly descr ibed in this paper for the first time.With the popularity of more advanced sensors,the massive data generated needs to be processed comprehensively before making the final decision for the reliable operation of the power system.On the basis of the three-dimensional information space,the data layer,the feature layer and the decision layer in the big data analysis structure correspond with the sensing collection layer,the data management layer,the application layer in the power system and a multi-layer data fusion processing scheme for large data and multi-source data is built.Secondly,In order to improve the efficiency of data fusion,the Hermite orthogonal base forward neural network algorithm is introduced.In order to cope with the large capacity of power data,the algorithm is parallelized under the framework of MapReduce,so as to realize the operation of subsequent large-scale data sets.Finally,the experimental platform is built on the basis of Hadoop.Through the experimental analysis of wind power prediction,the historical monitoring data of the wind farm is used as the source data,and the data are divided into several groups of different data to carry out the experiment.The comparison of the experimental results shows that the fusion algorithm proposed in this paper has obvious advantages over the traditional algorithm in power prediction accuracy and data processing efficiency.In this paper,the research on the big data fusion of electric power is of great significance to the processing and analysis of the big data of the power grid in the future.It also points out a way to walk for the power system of the big data age.