An Intelligent Optimised Estimation of the Hydraulic Jump Roller Length

In this paper, we address a problem in the field of hydraulics which is also relevant in terms of sustainability. Hydraulic jump is a physical phenomenon that occurs both for natural and man-made reasons. Its importance relies on the exploitation of the intrinsic energy dissipation characteristics and on the other hand the danger that might produce on bridges and river structures as a consequence of the interaction with the large vortex structures that are generated. In the present work, we try to address the problem of estimating the hydraulic jump roller length, whose evaluation is inherently affected by empirical errors related to its dissipative nature. The problem is approached using a regression model and exploiting a dataset of observations. Regression is performed by minimising the loss function using ten different black-box optimisers. In particular, we selected some of the most used metaheuristics, such as Evolution Strategies, Particle Swarm Optimisation, Differential Evolution and others. Furthermore, an experimental analysis has been conducted to validate the proposed approach and compare the effectiveness of the metaheuristics.