Life-Cycle Oriented Methods for Structural Analysis and Design of Bridges

Research focus

The Peer review has evaluated this group as Excellent

The objective of this research topic is to develop computational methods for RC and PC structures and to the refine some special numerical techniques (improved secant and tangent methods) to make them suitable to deal with both the mechanical and geometrical nonlinearities typical of bridge structures. A number of problems regarding frames, plates and shells have been studied, and there have been significant applications of the research results to the design of medium- and long-span bridges. These analyses pointed out the pros and cons of different approaches. Obviously, the more refined the analysis, the more accurate and complete the input data which is required (constitutive laws, load combinations, distortions and imperfections, etc.). Consequently, in design attention should be shifted from structural analysis to structural synthesis. With this in mind, great attention was given to Strut-and-Tie Models for RC structures. The use of optimisation techniques for an objective definition of the load path of the forces made it possible to formulate a numerical tool suitable for actual design problems. Several methods and algorithms based on the artificial intelligence tools (neural networks, genetic algorithms, fuzzy logic, cellular automata) were developed for the solution of various structural problems. The lack of information on many critical input data necessary for the analyses was overcome by assuming that the structures are intrinsically affected by uncertainties. A numerical approach for evaluating the structural reliability of concrete structures at the serviceability and ultimate limit states has been developed by using a probabilistic approach for the aleatoric part of the uncertainties and a fuzzy approach for the part characterized – for example - by a lack of knowledge (epistemic part of the uncertainties). At present special attention is devoted to RC structures exposed to severe environmental conditions, where the structural behavior must be considered as time-dependent, mainly because of the progressive deterioration of the mechanical properties of the materials, which impairs the capability of the structure to bear the applied loads. The aim is to develop concepts, techniques and methods for in-time structural analysis, for the evaluation of the service life, for the optimization of the design and for the planning of maintenance activities of damaged structural systems. Attention is also focused on concrete structures exposed to fire and to aggressive environments, but the research currently in progress can be easily extended to cover any structural type and any source of damage. In the context of durability analysis and lifetime assessment of concrete structures subjected to diffusive attacks from external aggressive agents, a novel approach has been developed wherein the diffusion process is modeled by using cellular automata and the mechanical damage coupled to diffusion is described by introducing specific laws for material degradation. Since the rate of mass diffusion usually depends on the stress state, the interaction between the diffusion process and the mechanical behavior of the damaged structure is also taken into account by modeling the stochastic effects of mass transfer. In addition, the randomness of the main structural parameters – be they related to materials properties (such as the diffusivity and damage rate) or to geometrical parameters (such as area and location of the reinforcement) is taken into account by using a Monte Carlo simulation. Therefore, the time dependence of the structural reliability is evaluated with respect to predefined performance levels and the results of the lifetime reliability analysis are used to select, among different maintenance scenarios, the most economical rehabilitation strategy, in accordance with the prescribed service life. Last but not least, another original methodology oriented towards the lifetime optimization of deteriorating structures has been developed, in this csae with the goal of finding design solutions which comply with the desired performance not only immediately after the completion of the structure but 96 also throughout the entire expected service lifetime. Of course, the effects induced by the many unavoidable sources of mechanical damage and by the rehabilitation interventions must be taken into account. The results obtained with this new approach shed light on the major role played by timedependent performance and by maintenance planning in working out the best design strategy. Many of the aforementioned analytical tools were recently applied to the design of several medium- and large-span bridges.

Departments

Dipartimento di Ingegneria Strutturale (DIS)

Professors

Full Professors
Pier Giorgio Malerba
Associate Professors
Fabio Biondini
Elsa Garavaglia