Road Surface Roughness Generation by Power Spectral Density in Bridge Design

Abstract

The effect of moving loads is important for the design of bridges because these loads cause vibrations when they are moving at high velocities. These loads vary in both space and time creating extra difficulty in the analysis. An impact factor is used in bridge design to take into account increased stresses above those due to static load to account for the dynamic effects of moving loads. The impact factor used in the Standard Specifications for Highway Bridges by AASHTO is given as I = 50 / ( L +125), where I is the impact factor with a maximum value of 30 percent, and L is the length in feet of the portion of the span that is loaded to produce maximum stress in the member. Impact factor given by AASHTO depends only on the span length of the bridge. When a moving load is traveling on a bridge, road surface profile, vehicle dynamics, weight and speed of vehicle, and the geometry of a bridge all play important part in affecting the impact factor. The overall objective of the research work was to study the collective effect of all the factors on the impact factor and compare it with the AASHTO values. Many researchers have studied the behavior of highway bridges under the action of moving loads from early 1950s. Some have studied the role of road surface roughness on the increased load while others have studied bridge deck behavior under a constant load or a train of loads moving across the bridge with different speeds. Dodds and Robson studied the road surface roughness. They showed that typical road surface may be considered as realization of homogeneous and isotropic Gaussian random process and proposed a new road classification method based on power spectral density function. Silva presented an analysis to evaluate dynamic effect of vehicles crossing the rough surface on highway bridge decks by a probabilistic model. He obtained results for heavy vehicles moving over rough straight reinforced concrete box girder highway bridge deck. It was found in this study that the dynamic effect can be as high as 90% on highway bridge decks due to the interaction of vehicle suspension flexibility with smooth surface pavements at usual vehicle velocities. Wang, Shahawy, and Huang studied dynamic response of highway trucks due to road surface roughness using the road classification method proposed by Dodds and Robson. They developed vehicle models for AASHTO H20-44 and HS20-44 trucks with seven and twelve degrees of freedom respectively. Impact factors for the suspension and tire forces for vehicle models running on different classes of roads were obtained at various speeds. Bridge decks can be considered as orthotropic plates for analysis. Free vibrations of rectangular orthotropic plates were studied by Jayaraman, Chen, and Snyder. Very little investigation has been undertaken about the forced vibrations of orthotropic plates under moving loads. In this research effects of the road surface roughness, vehicle dynamics, the weight and the speed of vehicles have been combined to calculate the increased load on the bridge decks. The bridge decks were analyzed by orthotropic plate theory and finite element methods to compare the results by using arrival time function data for the nodal points along the moving path of the AASHTO HS20-44 truck load. In this paper a portion of the research is presented. It is intended to show here how the road surface roughness profiles are generated by using power spectral density.