Analyses and modeling of track transition site responses from instrumented bridge approaches

Railroad bridge approaches, which connect open track to bridge structures, commonly experience recurring defects, such as broken superstructure components, excessive deformation, and ballast degradation that may require frequent maintenance. Such deformations or simply referred to as "bumps at the end of the bridge" are often associated with dynamic wheel loads and impact forces that are caused by stiffness variation and inevitable differential settlement accumulation between a railroad bridge and track. Although many studies have investigated this bridge approach problem, a permanent effective solution is yet to be agreed upon among researchers. This Ph.D. dissertation aims to contribute to the existing knowledge of the bridge approach problem through extensive analyses of field data and numerical modeling aimed to investigate mechanisms associated with the bump phenomenon. Multidepth deflectometers (MDDs) and rail strain gauges were installed as field sensors at three different bridge approaches on Amtrak's Northeast Corridor (NEC). The MDD sensors were used to collect transient vertical displacements and settlement accumulation trends of up to five individual track substructure layers for an investigation anchor depth of 3.05 m. (10 ft.) in the track substructure. Rail strain gauges installed at the same locations provided both the dynamic wheel loads and the tie reactions under train loading. Upon monitoring the performances of these NEC bridge approaches for over a year, the ballast layer was determined as the primary contributor to differential movement and settlement observed in these track transition zones. Stone blowing and polyurethane injection remedial measures were applied to the two bridge approaches to stabilize the ballast layer in these locations. The application of these mitigation efforts was achieved without disturbing the already placed MDD and rail strain gauge sensors. The effectiveness of these mitigation measures was evaluated by comparing the trends and magnitudes of the instrumentation data collected before and after the remedial measures. The results of the data analyses indicated that tie-ballast gaps occurred commonly at the bridge approach locations. The amount of this gap observed underneath the concrete crossties contributed significantly to excessive transient responses commonly collected at the bridge approach locations. A numerical study was conducted using a state-of-the-art ballast particle image-aided discrete element method (DEM) to investigate the impact forces generated in the presence of these tie-ballast gaps and evaluate their effects on ballast support conditions and individual ballast particle movements. In addition, using the polyhedral particles in the BLOKS3D DEM program, numerical simulations were conducted to study the micromechanical interactions of smaller sized ballast aggregate particles used in the stone blowing remedial measure, which essentially fill any tie-ballast gap and enhance tie support conditions to minimize ballast particle movements under dynamic loading. The results of the numerical study indicated that increasing amounts of gaps underneath ties resulted in larger impact forces between the ballast surface and caused larger particle movements. These particle movements were primarily permanent in nature in the transverse direction and displayed a heaving trend in the vertical direction. The heaving trend of the particles inferred to particle degradation that might occur in the field under impact forces. The insertion of small-sized particles for stone blowing simulations decreased this heaving movement due to better force distribution due to the increased tie ballast contact surface.U of I OnlyAuthor requested U of Illinois access only (OA after 2yrs) in Vireo ETD syste