Mass Data Storage Platform For Smart Grid
Huahui Lyu,Ping Li,Yuning Xiao,Hongjie Qian,Bin Sheng,Ruimin Shen
DOI: https://doi.org/10.1109/PIC.2016.7949557
2016-01-01
Abstract:With the continuous development of power systems, the scale of power grid is increasingly expanding, and the structure of it is becoming more and more complex. Traditional electrical network management cannot meet the current requirements. In this paper, we use load forecasting as a starting point, carrying on the exploration regarding heterogeneous multi-source data of the smart grid and doing research on short-term load forecasting. In the process of building the application platform of smart grid, we propose some solutions to solve these problems. In the study of heterogeneous multi-source data fusion, we proposed a fusion method of grid data based on CIM/XML after comparing the current method and do numerical simulation on our application platform of smart grid to test its feasibility. At the same time, we also do a lot research on the current situation of load forecasting. After analyzing the advantage and disadvantage of some normal forecasting methods, for example, the peer-fold ratio method, the ration smoothing method, grey prediction and variation coefficient method, we developed two comprehensive approaches, the method of weighted mean and the ratio method based on grey prediction, which will increase the accuracy, general applicability and stability of load forecasting. Through these studies and efforts, we build a friendly and visually interactive application platform, which will lay the foundation of further development about the application platform of smart grid.