Power Demand Data Analysis and Recovery for Management of Power Distribution Systems

Efficient anomaly detection and repair method is an important means to solve the problem of time series data quality. Current pair research on the quality of time series data mainly focuses on the lack of attribute value, data anomaly and data disorder, but the phenomenon of data loss is very common in the industry. This work presented a solution for power demand measurement analysis and recovery for the management of power distribution systems. The developed method and the algorithm are extensively evaluated and validated based on the simulation experiments for different test cases and the numerical results demonstrated that the proposed missing data recovery method can perform efficiently to recover the missing data for the improvement of data quality to enhance the operation of power distribution system operations. The solution can be further extended to different application domains to improve the data quality of advanced intelligent functionalities.