| 作成者 |
|
|
|
|
|
| 本文言語 |
|
| 出版者 |
|
|
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 号 |
|
| 開始ページ |
|
| 終了ページ |
|
| 出版タイプ |
|
| アクセス権 |
|
| Crossref DOI |
|
| 権利関係 |
|
| 概要 |
There have been many empirical equations in studies on flow over Piano Key Weir (PKW), but these equations are only suitable for each research model. Thus, there cannot be a correct general equation f...or all cases. Additionally, machine learning is a research method that can gather all the data to build a training model. The more data, the more accurate the prediction. Moreover, it is also a solution that tends to suit current developments. Currently, there are many machine learning algorithms, each with advantages and disadvantages. In particular, the Support Vector Machine (SVM) algorithm is also a regression algorithm with high prediction performance. This study analyzed the factors affecting the flow over type-A PKW according to Pi theory. From there, it established the prediction objective function in machine learning and then applied the SVM algorithm for prediction. The results indicated that the Medium Gaussian SVM model has good predicting performance, comparing the predicted values and the measured values showed a very high correlation coefficient (R^2 ＝ 0.97), other statistical indicators were very close to the ideal point (MSE = 0.001; RMSE = 0.033; MAE = 0.025). Furthermore, the largest percentage error was only 8.7%. This demonstrated that the SVM algorithm is suitable for studying and predicting flow characteristics over type-A PKW.続きを見る
|
Mendeley出力
