| 102D1 |
|---|
| A Prediction Method of Typhoon Damage of Electric Power Systems in Kagoshima Prefecture via RBF Networks |
| AUTHORS |
| Hitoshi Takata, Hirofumi Nakamura, Tomohiro Hachino, Nikio Matsuyama (Kagoshima University) |
| ABSTRACT |
| Kagoshima Prefecture suffers every year from typhoons, which often damage to power systems very badly and cut off electricity. They sometimes give a bad effect on our social life and inflict a considerable loss to the electric power company. In order to resume normal electricity services as soon as possible, it is necessary to predict accurately the damage which may be caused by typhoons. This paper proposes a method for predicting the typhoon damage to power systems by using a radial basis function networks which use a multiquadric function as a base function. From the weather information of wind velocity and the typhoon progress, this method enables us to predict the degree of damage to electric distriction lines and poles. The effectiveness of this prediction method is demonstrated by applying it to the actual data. Simulation results show that this prediction method is better than those of liner regression model and neural network. |
Back to Program
Time Table
Top of This Site
Annual Conference 2002