Applying Particle Swarm Optimization algorithmon normal depth calculation of horse-shoe section tunnel

Zhang Kuandi,Wang guangqian,Lü Hongxing,Hong Cheng
DOI: https://doi.org/10.3969/j.issn.1674-8530.2011.01.012
2011-01-01
Abstract:Owing to the fact that the calculation formulas of normal depth for free flow in horseshoe section of tunnel and drainage culvert are not expressed by explicit function in hydraulics.By mathematical transformation of normal depth equation of horse-shoe section tunnel,a model of nonlinear constrained optimization for calculating the normal depths of standard Ⅰ-type and Ⅱ-type horse-shoe section tunnel was established.In order to accelerate the convergence rate of the algorithm and improve the searching ability of particle,an improved Particle Swarm Optimization algorithm was presented.The dynamic inertia weight was changed in every loop according to the particle′s positions and the distance between the optimization particle.Error analysis and a computed illustration using the new method indicate that it is much more applicable,precise and simple than traditional methods for calculation of normal water depth.At the same time,the correctness and validity of the new method was demonstrated.So it provided a new tool for obtaining normal depth of open channel with horse-shoe section problem.
What problem does this paper attempt to address?