@article{oai:kanazawa-u.repo.nii.ac.jp:00008684,
 author = {小松﨑, 俊彦 and 岩田, 佳雄 and 本江, 哲行},
 issue = {784},
 journal = {日本機械学会論文集C編 / Nihon Kikai Gakkai Ronbunshu, C Hen / Transactions of the Japan Society of Mechanical Engineers, Part C},
 month = {Jan},
 note = {It is important to identify the information of the external force acting on the structures in order to address the vibration related issues. More specifically, the identification of the impact point would offer advantages in some sports training where the recognition of the hitting accuracy is thought to be effective for the skill improvements. In this study, a kind of the artificial neural networks known as Radial Basis Function Network is introduced for the estimation of the impact point in structures based on the measured acceleration responses obtained by relatively small number of pre-determined force inputs. The force input location is predicted for both one and two-dimensional problems where the effects of the number of sample points used for the network learning and the size of the input vector on the estimation accuracy are investigated. © 2011 The Japan Society of Mechanical Engineers},
 pages = {4521--4533},
 title = {放射状基底関数ネットワークを用いた打撃点の推定にする研究},
 volume = {77},
 year = {2011}
}