|
Multi-scale Fatigue Damage Prognosis of Orthotropic Steel Decks of a Cable-stayed Bridge under Moving Vehicles
You-Lin Xu, The Hong Kong Polytechnic University Chair: Yuan-Sen Yang Time: 2019/12/18 13:20-14:00 (101C, 1F, TICC) |
Abstact:
Orthotropic steel deck (OSD) is widely used in long-span cable-supported bridges because of its high capacity-to-weight ratio, high quality prefabrication, and easy on-site assembly. However, the fatigue failure of deck-to-rib (DTR) welded joints of OSD under moving vehicles has been observed recurrently, which poses a threat to the bridge functionality and safety. The prognosis of vehicle-induced fatigue damage of DTR joints of OSD is therefore imperative for the optimal design, safety assessment and maintenance of the bridge. However, this is a very challenging task in consideration of huge size of a long-span cable-stayed bridge, stochastic traffic flow, coupled vehicle-bridge dynamics, complicated local weld geometric configuration, OSD and pavement interaction, and complex fatigue mechanism.
In this regard, this paper presents a framework for vehicle-induced fatigue damage prognosis of DTR welded joints of OSD in a long-span cable-stayed bridge. The framework considers hourly traffic flow/loading simulations other than currently-used daily traffic flow/loading simulations so that the both seasonality and growth trend effects on fatigue accumulation can be considered appropriately. Another feature of the framework is the development of multi-scale finite element (FE) model for a cable-stayed bridge so that dynamic stresses at DTR welded joints can be computed feasibly and accurately. In the multi-scale FE model, solid elements are used to simulate DTR joints of OSD interacted with pavement (local model). Shell elements are used to simulate the most vulnerable girder segment (intermediate model). Beam/truss elements are used to simulate the rest part of the bridge (global model). With the appropriate coupling of the global, intermediate, and local models, the multi-scale model can accurately capture not only the global structural behavior in terms of displacement and acceleration but also the local structural behavior in terms of strain and stress simultaneously. The other features of the framework include fully coupled vehicle-bridge dynamics, mesh-insensitive equivalent stress responses at a DTR joint, a substitutive model for equivalent stress responses, OSD and pavement interaction, asphalt pavement temperature effect, fatigue test-generated S-N curve, and fatigue damage prognosis.
The Stonecutters cable-stayed bridge in Hong Kong is then taken as a case study to evaluate the feasibility of the proposed framework and to manifest the effects of pavement roughness, asphalt temperature, and transverse locations of vehicles on hourly and annual fatigue damage accumulation. The results show that the proposed framework is feasible and effective and that the fatigue damage of a DTR joint will be underestimated if the time-variant temperature of asphalt pavement and the degeneration of road surface condition are not considered but the fatigue damage will be overestimated without consideration of variable transverse locations of vehicles.
Orthotropic steel deck (OSD) is widely used in long-span cable-supported bridges because of its high capacity-to-weight ratio, high quality prefabrication, and easy on-site assembly. However, the fatigue failure of deck-to-rib (DTR) welded joints of OSD under moving vehicles has been observed recurrently, which poses a threat to the bridge functionality and safety. The prognosis of vehicle-induced fatigue damage of DTR joints of OSD is therefore imperative for the optimal design, safety assessment and maintenance of the bridge. However, this is a very challenging task in consideration of huge size of a long-span cable-stayed bridge, stochastic traffic flow, coupled vehicle-bridge dynamics, complicated local weld geometric configuration, OSD and pavement interaction, and complex fatigue mechanism.
In this regard, this paper presents a framework for vehicle-induced fatigue damage prognosis of DTR welded joints of OSD in a long-span cable-stayed bridge. The framework considers hourly traffic flow/loading simulations other than currently-used daily traffic flow/loading simulations so that the both seasonality and growth trend effects on fatigue accumulation can be considered appropriately. Another feature of the framework is the development of multi-scale finite element (FE) model for a cable-stayed bridge so that dynamic stresses at DTR welded joints can be computed feasibly and accurately. In the multi-scale FE model, solid elements are used to simulate DTR joints of OSD interacted with pavement (local model). Shell elements are used to simulate the most vulnerable girder segment (intermediate model). Beam/truss elements are used to simulate the rest part of the bridge (global model). With the appropriate coupling of the global, intermediate, and local models, the multi-scale model can accurately capture not only the global structural behavior in terms of displacement and acceleration but also the local structural behavior in terms of strain and stress simultaneously. The other features of the framework include fully coupled vehicle-bridge dynamics, mesh-insensitive equivalent stress responses at a DTR joint, a substitutive model for equivalent stress responses, OSD and pavement interaction, asphalt pavement temperature effect, fatigue test-generated S-N curve, and fatigue damage prognosis.
The Stonecutters cable-stayed bridge in Hong Kong is then taken as a case study to evaluate the feasibility of the proposed framework and to manifest the effects of pavement roughness, asphalt temperature, and transverse locations of vehicles on hourly and annual fatigue damage accumulation. The results show that the proposed framework is feasible and effective and that the fatigue damage of a DTR joint will be underestimated if the time-variant temperature of asphalt pavement and the degeneration of road surface condition are not considered but the fatigue damage will be overestimated without consideration of variable transverse locations of vehicles.
