Long-term Icing Level Forecasting Model Based on Support Vector Machine for Electric Power Grid
Electric power grid long-term ice disaster forecasting can provide valuable leading time for fighting against ice disaster.However,the long-term ice disaster is influenced by many factors,such as sunspot,atmospheric circulation,air temperature,precipitation,and so on.And the record about electric power grid ice disaster is usually very short.It is hard to build proper long-term ice disaster forecasting model.In this paper,several key factors are chosen by ice disaster causes analysis and correlation analysis.Then based on the powerful small sample learning ability of Support Vector Machine,a long-term ice disaster level forecasting is constructed for Hunan electric power grid.The results show that this model has gained very good performance and can be a powerful tool for long-term ice disaster forecasting.
Electric power grid ice disaster level long-term Support Vector Machine
GUO Jun LU Jiazheng YANG Li XU Xunjian FENG Tao JIAN Zhou
State Key Laboratory of Disaster Prevention and Reduction for Power Grid Transmission and Distribution Equipment,Changsha 410129;State Grid Hunan Electric Power Company Disaster Prevention and Reduction Center,Changsha 410129
国际会议
重庆
英文
1-4
2017-09-25（万方平台首次上网日期，不代表论文的发表时间）