RICERCA ITALIANA     

Health assessment and monitoring of civil engineering structures through advanced dynamics methods

Università degli Studi di Perugia

Abstract

The research project is aimed at the optimal resource allocation for the maintenance of civil engineering structures through the improvement of the knowledge of their service behavior and through the development of health monitoring and assessment techniques based on dynamic diagnostic methods.
The optimization will be achieved by merging the current technical-scientific knowledge with new methodologies that will lead to significant overall innovation. The general objectives are:
a) developing methodologies and equipments for estimating the structural integrity by monitoring the dynamic response, with main focus on highway and railway infrastructures and buildings belonging to the historical heritage;
a) providing unambiguous methods for damage assessment in order to make the design, construction and management process of civil engineering structures economically more effective for the whole service life;
c) collecting, recording, refining, disseminating, and promoting the specific knowledge in the field, in order to have the construction industry benefit from this information.
The complexity of the fixed targets requires the synergistic effort of a wide range of expertise in the civil engineering field.
For this reason the study will be carried out with the joint contribution of researchers belonging to 5 different units, some of them already collaborating in previous national research projects like the PRIN 2004 “VINCES - Vibrations in Civil Engineering Structures: source of damage and discomfort, diagnostic and safety assessment tool”. The partial and global results obtained will be reflected into significant social and economical benefits since they will support the construction industry in order to achieve high levels of expertise and efficiency, optimizing the construction and maintenance costs, improving quality and lifetime of structures and infrastructures. <<<

Principal Investigator

Annibale Luigi Materazzi Università degli Studi di PERUGIA

Research Objectives

The main goal of the present proposal is to improve the knowledge of the service behaviour of existing structures, with special reference to the infrastructures and buildings with historical and artistic significance in order to immediately identify the onset of damage that can prejudice safety and to give useful information for their maintenance, within the general framework of the optimal resource allocation.
This goal can be obtained through the following intermediate achievements:
A) developing methodologies and equipments for the periodical or permanent, monitoring of “strategical” buildings;
B) providing a method for the identification and location of damage and for the assessment of the severity of the damage of existing buildings, based on the analysis of their dynamic response to both artificial and ambient excitation.
C) collecting, recording, refining, disseminating, and promoting the specific knowledge in the field, in order to have the construction industry benefit from this information.
To obtain each one of the intermediate achievements, the following tasks are planned:
Intermediate achievement A:
- Collect of literature and experimental data concerning the equipment available for the design of new acquisition systems of the structural dynamic response, with particular reference to innovative technologies based on remote control systems and connected through “wireless” technology and data transmission systems based on wide band technology, using the Internet;
- Design of prototypes of monitoring systems;
- Selection of existing structures to use as case studies for the experimental tests;,
- Application of the developed monitoring techniques to the selected case studies and execution of pilot investigations.
Intermediate achievement B:
- Critical analysis of the current methodologies for assessing the structural integrity by dynamic methods;
- Development of innovative methodologies for the damage assessment through dynamic methods capable to overcome the limits inherent in the existing procedures;
- Laboratory experimental tests, to validate the proposed methodologies;
- Application of the techniques for damage identification to the previously defined case studies.
Intermediate achievement C:
- Costs-benefits analysis;
- Interaction with Professional Associations, Construction Industry and Institutions managing big infrastructure networks (that are not formally involved in this Research Project), also through the organization of Conferences and Workshops;
- Development of guidelines for assessing the structural integrity using dynamic methods.
Each one of the previous tasks might coincide with a part or a whole working stage of many Units. <<<

Timescale

24 months

National and international background

During their service life, civil engineering structures are subjected to damage due to the effects of the changes of the material properties (consequence of aging and chemical-physical interaction with the environment) and of the applied forces (both ambient and anthropic forces). Often these phenomena can realize simultaneously, giving rise to particularly unfavorable conditions.
The corresponding damage can lead to the decrease of the structural safety, can prejudice the structural service-ability and can reduce the duration of the service life.
The economical consequences of these phenomena are relevant in the case of infrastructures systems but can be significant, although not always easily quantifiable, also in the architectural heritage area.
It is very well known the case of the maintenance of the highway and railway bridges, that involves the allocation of very relevant resources for the collectivity.
On those themes have been carried out important research projects, like the BRIME (Bridge Management in Europe) and SAMCO (Structural Assessment, Monitoring and Control) funded by the European Union.
A wide report on the state of the art on these topics can be found in Sohn et. al (2003) [1].
The current tendency is to keep under observation the “sensitive” structures in order to suddenly identify the onset of the damage and being able to intervene before the damage consequences require too costly repairs.
One of the first systematic application of monitoring systems, documented in the literature, was carried out during the thirties on the Golden Gate Bridge and on the Bay Bridge of San Francisco, in order to study the dynamic response with the purpose of foresee the possible consequences of earthquakes (Carder (1937) [2]).
More recently the development of transducers systems having good performances and relatively low cost, together with the spread of acquisition systems with high velocity and elevated number of channels made possible the installation of permanent monitoring systems in many of the new long span bridges in Japan, North America and Europe (Brownjohn et al. (2005) [3]). Among them can be recalled the suspension bridge over the Humber in the United Kingdom, that from 1984 to 1998 has been the longest bridge of the world (Brownjohn et al. (1987) [4]), the suspension bridge over the Great Belt in Denmark (Jensen, (2006) [5]) and the Akashi-Kaikyo in Japan (Fujino e Kashima (2006) [6]), that is currently the longest existing bridge, with a maximum span length of 1991 m.
The current tendency is towards using the obtained data to identify eventual changes in the dynamic response, aimed at deducing information on the structural integrity of the relevant cross sections.
The identification of the damage in the bridges and, more in general, in the civil engineering structures through dynamic tests is however a complex task, as the parameters involved in the investigation process (natural frequencies of vibration, modal shapes, damping) depend on the overall structural stiffness and are therefore influenced in reduced proportion by the damage effects, that are usually localized. Often the variation of the dynamic behavior is so small that can be confused with the effect due to the changes of the ambient temperature, the humidity content and the masses added during the service life.
Moreover, the correlation between the stiffness of the structural elements and the mechanical strength of the cross sections, in most real cases, is weak and therefore the safety estimate is difficult. It must be pointed out that some investigation techniques require the knowledge of the dynamic properties of the undamaged structures, information that are rarely allowable.
A wide bibliographic review on the topic can be found in the Special section: Phase I of the IASC-ASCE Structural Health monitoring benchmark [7].

These themes have also been treated in the two editions of the Workshop: Vibration problems in civil structures and mechanical constructions” [8,9], that took place in Perugia, in 2001 and 2004, respectively.
According to the first studies on the topic (Lifshitz e Rotem, (1969) [10]) the natural frequencies of vibration can be assumed as reliable estimators of the damage, while more recent studies demonstrated that this is not always true (Spyrakos et al. (1990) [11]).
Also the modal shapes do not always represent a good indicator of the presence of the damage, as it was shown by Salawu and Williams [12] in 1994 and by Farrar and Doebling [13] in 1997, who carried on dynamic tests on the I-40 bridge over the Rio Grande in Albuquerque, USA.
More recent studies focused the attention on the use of advanced techniques for the analysis of the dynamic response based on statistical analysis, on proper damage indexes, on model updating, on neural networks , etc.
Standard methods of experimental modal analysis, like the Modal Assurance Criterion (MAC) and the Coordinate Modal Assurance Criterion (COMAC) have been used, among the others, by Fryba and Pirner in 2001 [11] to evaluate the damage, by comparing the undamaged and damaged structural response. In particular the COMAC method was used with success by the Authors to assess the effectiveness of the repair works in a three span prestressed concrete segmental bridge. The dynamic behaviour of the undamaged bridge was determined from a twin bridge located near the damaged one, which was taken as reference.
Brincker et al. (2001) [15] reported their experience of 15 dynamic tests carried out on a bridge in Switzerland, razed and replaced by a new bridge. The artificially realized a state of progressive damage. The technique used to determine the damage, called Enhanced Frequency Domain Decomposition (FDD), was based on the determination of eigenfrequencies, eigenvectors and damping changes. The Authors, neglecting the influence of environmental factors, such as temperature, on the modal properties, believed that they could determine the presence of damage even for small changes of the modal parameters.
If the damage consists in local cracks, the most part of references that can be found in the literature are about small size steel structures and are based on the hypothesis that cracks are always open (Patil and Maiti (2003) [16], Lin et al. (2002) [17], Kim and Stubbs (2003) [18]).
On the contrary, just a few studies concern the behavior of damaged prestressed or reinforced concrete structures, which is more complex than the one of structures made of homogeneous materials. The main reason of this difficulty is that the cracks, crossed by the steel reinforcements, open and close alternatively during the vibration, giving rise to changes in the dynamic response of the damaged structural element.
In 2000 Maeck et al. [19] described two methods for the determination of the stiffness decrease of structural concrete beams due to damage. The first method uses an updating algorithm based on the stiffness characteristics, which are damage-dependent, and on the comparison between the undamaged and the damaged structural response. It is based only on the eigenfrequencies and, therefore, it cannot detect asymmetric damage in a symmetric structure. The second one, called Direct Stiffness Calculation, is able to detect and localize damage through the modal shape determination.
The non-linearity of damaged reinforced concrete structural elements was recently studied jointly in the time-frequency domain by Neild et al. (2003) [20] and by Owen et al. (2001) [21]. A finite element model for damaged reinforced concrete elements based on the theory of the Fracture Mechanics was proposed by Saavedra &amp; Cuitiño (2001) [22]. Also Petryna e Krätzig (2005) [23] proposed a procedure for the damage evaluation based on the strip modeling of structures made of prestressed or reinforced concrete.
Within this general framework, the local Unit of Milan has carried out extensive campaigns of dynamic experimental tests on bridges and historical and monumental buildings with the main aim of the dynamic identification, experimental modal analysis and development of modal identification procedures and methods for the updating of uncertain parameters of structural models. In particular, theoretical-experimental investigations have been performed on the dynamic behavior of cable-stayed bridges (Gentile and Martinez Y Cabrera (1997) [24], Gentile and Martinez Y Cabrera (2004) [25]), of masonry bell towers (Limongelli and Pezzoli (1994) [26]) and on steel floors with different damage levels (Gentile (2001) [27]).
The local Unit of Perugia has performed general studies on the sensitivity of the dynamic methods for the identification of damage in reinforced concrete bridges (Breccolotti et al. (2004) [28]) and carried out experimental tests using representative prototypes of reinforced concrete beams, subjected to artificial damage and studied using excitation provided by an instrumented hammer and electrodynamic shaker (Breccolotti e Materazzi (2005) [29]).
The local Unit of Palermo has studied structures subjected to ambient loads, modeled as stochastic processes, taking into account the uncertainties on geometrical and mechanical parameters of the structures (Caddemi and Di Paola (1997) [30], Di Paola and Pirrotta (1999) [31], Di Paola et al. (2004) [32]).
The local Unit of L’Aquila has pointed the attention on the monitoring of large structures, in particular on monumental buildings (Valente (2003) [33], Antonacci et al. (2002) [34]) and the related problems of analytical modelling of the dynamic response. Special attention has been devoted to the behavior of cables for cable-stayed bridges under wind excitation (Gattulli and Lepidi (2003) [35], Gattulli et al. (2002) [36]).
The local Unit of Florence dealt with the monitoring of monumental masonry buildings, applying advanced techniques for the analysis of the response to the study of their maintenance conditions (Bartoli et al. (2002) [37], Chiostrini and Facchini (2002) [38]). <<<