Abstract :
Wind generation is one of the fast growing and introduced resources among renewable energies through worldwide including Japan. As Japan, on the other hand, is an island country surrounded by ocean, the topography suitable for wind generation is limited for the on-shore. Therefore, based on the wind map of up to 2030, it is expected that new introduction suitable for wind power generation will be more on off-shore. For this reason, it is very important to determine the wind characteristics of the candidate area for installing wind generation, however in most cases of off-shore installation, existence of weather condition data is poor and it may need lots of time and cost for collecting such weather condition data as a new pin-point information. In this study, the goal of this research is to project a wind speed of an unseen area (where its weather condition data are not available) by mapping the modularized Artificial Neural Network (SOM: Self-Organization Map) of seen areas (where their weather condition data are available) around the target area. By learning the correlation between modularized ANNs of seen and unseen areas, the results of these temporal and spatial projections will be the prediction of wind speed of target place. It is believed, by the help of the proposed technique, a huge amount of time and cost will be saved in the selection of installation point of off-shore wind power generation. Moreover, it is intended to contribute to the introduction increase in the amount of wind power.
Keywords :
"Wind speed","Mechatronics","Artificial neural networks","Observatories"
