Dimensionamento de bueiros de talvegue: uma análise comparativa entre as metodologias do DNIT e da FHWA

Abstract

Road infrastructure is fundamental to the country's socioeconomic development, and its durability depends directly on an efficient drainage system. The proper sizing of highway culverts is a critical step in this process, and the available methods for this purpose present distinct approaches that may result in divergent designs. This study aims to evaluate the differences between the Critical Flow Method, recommended by the Road Drainage Manual of the Brazilian National Department of Transport Infrastructure (DNIT), and the Hydraulic Design of Highway Culverts method (HDS-05), developed by the Federal Highway Administration (FHWA), in the sizing of highway culverts, using as a case study five culverts belonging to the executive project of the Contorno Noroeste de Pato Branco, located in southwestern Paraná, Brazil. For each culvert, hydrological studies were conducted, including determination of the time of concentration using the Kirpich equation, calculation of rainfall intensity using the intensity-duration-frequency equation for Pato Branco, and estimation of the design flow using the simple or corrected Rational Method. The DNIT dimensioning followed the Critical Flow procedure, with verification of critical discharge, flow regime, and design velocity. The HDS-05 dimensioning was performed using the HY-8 software, with analysis of inlet and outlet controls and identification of the minimum section that meets the established hydraulic criteria. The results demonstrated dimensional convergence between the methods in four of the five culverts analyzed. The DNIT Critical Flow Method proved consistently conservative, producing safe designs in all cases, but without providing information on the governing hydraulic control mechanism, the effective safety margin, or the actual flow conditions inside the culvert. The HDS-05, by performing a complete energy balance and distinguishing between inlet and outlet controls, proved analytically more complete, leading to designs better adjusted to real hydraulic conditions. It is concluded that the adoption of the HY-8 software as a complementary tool to the DNIT method is technically advisable, especially for longer culverts, low slopes, or high design flows.