Multi-Regression Model for Peak Load Forecast in Demand Side like University Campus
The load/demand forecast of the electric power has been studied extensively in the world. In many case, the demand forecast of the electric power has been studied in power utilities, not in the demand side. The demand forecast in limited area is more difficult than in utilities. Although the contents of a load are limited, the behavior of loads is not expectable in advance. In the era of deregulation, however, the management of customers’ load became more important. The accurate load forecast brings economical benefits to customers. Recent efforts to promote renewable energy requests more “flexible load to mitigating the intermittency of the wind/solar power. In this paper, the multi-regression model for a peak load forecast in demand side is studied. The motivation of this study is the optimal operation of BESSs (battery energy storage systems) such as NAS battery in demand side. Five explanatory variables are selected in the multi-regression model. Since those variables are not quantitative in nature, they are converted to numerical variable using simple tables. Local information can be reflected using those tables graphically and schematically. This approach is well accepted by energy managers in sites. Numerical tests demonstrate the robustness and the accuracy of the proposed model.
Load Forecasting Power Demand Quantification Theory Multi-Regression Model Battery Storage Plants
Jun Shimizukawa Chao-Yuan Chen Kenji Iba Yusuke Hida Ryuichi Yokoyama Kouji Tanaka Kuniaki Yabe
Department of Electrical Engineering, Meisei University, Hino, Tokyo 191-8506 Japan Graduate School of Electrical Engineering, Meisei University Graduate School of Environment and Energy Engineering, Waseda University, Shinjuku-ku, Tokyo 169-005 Tokyo Electric Power Co. Inc., Chiyodaku,Tokyo, Japan Tokyo Electric Power Co. Inc., Chiyodaku, Tokyo, Japan
国际会议
The International Conference on Electrical Engineering 2009(2009 电机工程国际会议)
沈阳
英文
1-6
2009-07-05（万方平台首次上网日期，不代表论文的发表时间）