Research On Processing Methods Of Multi-source Heterogeneous Data For Power Dispatching Control System

With the further implementation of the smart grid construction and the extensive use of smart sensing equipment,the data volume of power system is explosive growth trend.Power industry inevitably entered the big data era.Power big data has broad application prospect,but it needs to face of some problems like huge amount of data,heterogeneous data and the scatter of data,etc.In order to realize the effective use of big data in dispatching control system,the multi-source heterogeneous data in this system need to be shared to establish a unified data model and panoramic data.It is of great significance for the study of multi-source heterogeneous data processing method and the data fusion technology in electric power dispatching control system.Firstly,based on the analysis of data characteristics,demand in dispatching control system and multi-source heterogeneous data processing methods,the panoramic uniform data model and the ETL-model are proposed to meet the demand of establishing a panoramic data.The thesis analyzes the system frame and workflow of the model.Secondly,aiming at the incomplete data appeared in the process of building the panoramic data,a missing data filling algorithm for uniform data model is proposed in panoramic dispatching and control system.The best parameters of the mean and covariance are estimated by improved chaos genetic optimization algorithm.Then,the missing data is estimated by improved Markov Chain Monte Carlo method based on the known data.The results show that the algorithm can get incomplete data parameters with fewer iterations,improve the estimated value accuracy of missing data to ensure the accuracy and completeness of the data.Finally,a string matching algorithm is proposed in multi-source heterogeneous data integration to solve the string matching problem in the process of multi-source heterogeneous data integration.In order to match string accurately and fast,the match rules are proposed based on the data set characteristics of dispatching control system.Meanwhile,the method of calculating matching degree is proposed to improve the accuracy and promote the building panoramic data.The effects are verified by the experimental simulation and the algorithm facilitates the regulation of multi-source heterogeneous data integration.