Predicting the accurate hydrodynamic pressure at plunge pool bottoms due to the impact of plunging high-velocity jets is essential in assessing the stability of the bed rock blocks and concrete slabs. In this context, the subsequent scour depth evaluation is essential. The regression-derived model of multiple nonlinear regression (MNLR) and two intelligent models of artificial neural network and adaptive neuro-fuzzy inference system are developed to predict the hydrodynamic pressure mean and hydrodynamic pressure fluctuations at flat and scoured plunge pool bottom. By running statistical analysis on a wide range of large-scale experimental data, it is revealed that, in general the intelligent models outperform the regression-derived equations of MNLR. The average values of RMSE and R² in the prediction of hydrodynamic pressure coefficients are improved to 0.054 and 0.87, respectively. Nevertheless, due to its simplicity the empirical equations based on MNLR model are accurate enough for engineering applications. These equations predict both the dynamic pressure mean and root mean square of pressure fluctuations at flat bottoms and scoured bottoms more accurately than the available equations for this purpose, with 70% and 90% of data within 20% range of discrepancy, respectively. An empirical equation of pressure fluctuation at SB is introduced for the first time in this study.

在评估基岩块和混凝土板的稳定性时，预测由于高速射流的撞击而导致的在水池底部的精确水动力压力至关重要。在这种情况下，随后的冲刷深度评估至关重要。建立了多元非线性回归（MNLR）的回归模型以及人工神经网络和自适应神经模糊推理系统的两个智能模型，以预测平坦和冲刷的水底池底部的水动力压力平均值和水动力压力波动。通过对各种大型实验数据进行统计分析，可以发现，一般而言，智能模型的性能优于MNLR的回归方程。RMSE和R ²的平均值在预测水动力压力系数时，分别提高到0.054和0.87。然而，由于其简单性，基于MNLR模型的经验方程对于工程应用足够准确。这些公式比为此目的可用的公式更准确地预测了平底和擦洗底部的动压均方根值和均方根值，分别有70％和90％的数据在20％的差异范围内。这项研究中首次引入了SB处压力波动的经验方程式。