컨텐츠 시작
학술대회/행사
초록검색
| 제출번호(No.) | 0583 |
|---|---|
| 분류(Section) | Poster Session |
| 분과(Session) | Control Theory / Optimization (SS-13) |
| 영문제목 (Title(Eng.)) |
[CANCELLED] Modeling swimming using with artifitial neural network and evaluation of results |
| 저자(Author(s)) |
Hamed Alavinejad1 Islamic azad University of Bojnourd1 |
| 초록본문(Abstract) | Neural Networks are capable because of having useful features empowers us for our modeling, Solving Nonlinear and Optimization issues. The purpose of this paper is a designed neural network of feedforward that Modeled with swimmers data is learned to simulate swimmers performance and use with evaluation factors like Correlation coefficient (R), Root mean square error (RMSE) and Mean absolute error (MAE) evaluate the network. In this paper case study is youth swim team of Bojnourd city, hence 80% of the data is for training neural network model and 20% of the data is for evaluation the network. To train the network several learning algorithms are used like BP, RBP, Scaled Conjugate Gradient and Levenberg-marquardt that learning algorithm levenberg-marquardt because of having Lowest possible differences in the physical model output was chosen as the best network. Finally, simulation results come in the graphs such as regression graph, simulation graph, training process graph and the efficiency function graph. |
| 분류기호 (MSC number(s)) |
00A06, 00A71, 00A72 |
| 키워드(Keyword(s)) | neural network, modeling, swimming, correlation coefficient |
| 강연 형태 (Language of Session (Talk)) |
English |