The experiment was conducted to compare the hydraulic openness by the buoyancy theory of the hydrostatic pressure and the hydraulic openness by the measurement, and the difference between the theory and the measured hydraulic openness was confirmed to be the effect by the downpull force. As a result of using the existing theoretical formulae to examine the downpull force of the buoyancy-type floating gate, it was found that the existing formulae were difficult to apply to the buoyancy-type floating gate. Therefore, this paper proposed the use of the multiple regression analysis and adaptive neuro-fuzzy inference system method for predicting downpull force coefficient of the buoyancy-type floating gate. The experimental results indicate that the downpull force coefficient decreased as the opening ratio increased. In addition, the downpull force increased as the opening ratio increased below the opening ratio of 0.300, however the downpull force decreased as the opening ratio increased from the opening ratio exceeding 0.300. Finally, prediction results by using the proposed relationship indicated that the errors are much smaller compared to the results of the existing theoretical formulae. In addition, the application of the adaptive neuro-fuzzy inference system method improved the prediction significantly, reducing the errors to the half of that by the existing theoretical formulae.
Read full abstract