1. School of Control and Computer Engineering, North China Electric Power University, Baoding 071003, China;2. College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China;3. Department of Electrical and Electronic Engineering, Institut Teknologi Brunei, Bandar Seri Begawan BE1410, Brunei;4. State Grid Henan Electric Power Company, Zhengzhou 450052, China

By taking into account the multi-dimensional, hybrid spatial and temporal characteristics of data from a power system and/or a power information system, a big data attribute reduction preprocessing method based on partial order reduction is presented. In the proposed method, the parallel advantage of MapReduce is employed and the independence of concurrent events is focused on, making it possible to meet the coverage requirements of big data attribute dimension and reduction in a power system. Finally, the proposed attribute reducing method is simulated by some examples including the monitoring data of a photovoltaic (PV) power generation system, the fault diagnosis data of a transformer, and real-time and reliability data of the communication system of an intelligent substation, with good performance observed through a Hadoop platform. This work is supported by National Natural Science Foundation of China (No. 51477151, No. 51407076), Hebei Provincial Natural Science Foundation of China (No. F2014502050), the Higher Education Scientific Research Project of Hebei Province (No. Z2013007), and Fundamental Research Funds for the Central Universities (No. 2015ZD28).