Vol.:(0123456789)
Journal of Civil Structural Health Monitoring
https://doi.org/10.1007/s13349-019-00369-0
123
ORIGINAL PAPER
A study on the concurrent influence of liquid content and damage
on the dynamic properties of a tank for the development
of a modal‑based SHM system
C. Rainieri
1
· D. Gargaro
2
· E. Reynders
3
· G. Fabbrocino
1
Received: 27 March 2019 / Revised: 12 September 2019 / Accepted: 26 November 2019
© Springer-Verlag GmbH Germany, part of Springer Nature 2019
Abstract
Remote detection of structural aging, degradation phenomena and damage due to hazardous events is critical to ensure safety
and reliability of civil or industrial structures. This is the motivation of the rapid development and increasing application of
fully automated structural health monitoring (SHM) systems in civil engineering. Modal-based damage detection currently
represents a popular approach for SHM of civil structures. In fact, it is a global method for damage detection and, as such,
the measurement locations are not required to be close to the damage. However, damage sensitive features defined in terms
of modal parameter estimates are also influenced by environmental and operational factors. Thus, neglecting this influence
might jeopardize the reliability of the technology. In this framework, the present paper investigates the potential of modal-
based SHM to detect earthquake damage at anchors of an atmospheric liquid storage tank. To this aim, the influences of bolt
loosening at supports and of liquid level on modal parameters are investigated. The ultimate goal of the present study is the
definition of criteria for the effective design and application of modal-based SHM to liquid storage tanks.
Keywords Modal properties · Liquid storage tank · Liquid level · Bolt loosening
1 Introduction
Liquid storage steel tanks are critical components of indus-
trial sites (petrochemical industries, refineries) as well as
strategic structures such as hospitals and other emergency
facilities. Degradation phenomena due to aging or severe
environment, and damage induced by hazardous events,
such as earthquakes, fire or explosions, may affect their per-
formance and safety [1]. Research studies on liquid-filled
tanks [2–6] represent the basis to understand their dynamic
and seismic behavior and for relevant codes and guidelines
(see, for instance, [7]) adopted to design, test and operate
metallic liquid storage tanks over the years. As a result of the
accumulated experience over some decades, liquid storage
tanks can currently be designed in a reliable way. However,
the need for remote and automated detection of damage or
degradation phenomena stimulated wide research efforts on
safety evaluation of liquid storage tanks. In fact, damage to
tanks might cause temporary problems, such as shortage of
water supply, as well as long-term issues, such as environ-
mental pollution, and soil and water contaminations. Visual
inspections, eventually in combination with nondestructive
investigations, and scheduled maintenance are the usually
applied approaches to damage detection and management.
However, the tank has to be taken out of service during the
inspections, and this might cause serious financial and safety
implications. This is why permanent structural health moni-
toring (SHM) systems are increasingly applied also to this
type of structures.
A wide range of structural and environmental param-
eters can be collected to identify damage at an early stage.
Fiber optic sensors are probably one of the most attrac-
tive technologies in the field [8–10], thanks to the excel-
lent sensing performance and electromagnetic immunity.
However, quasi-static monitoring shows some limitations
* C. Rainieri
carlo.rainieri@unimol.it
1
Structural and Geotechnical Dynamics Laboratory StreGa,
DiBT Department, University of Molise, Viale Manzoni snc,
86100 Campobasso, Italy
2
S2X s.r.l, Piazzale Scarano 6, 86100 Campobasso, Italy
3
Deparment of Civil Engineering, Structural Mechanics
Section, University of Leuven (KU Leuven), Kasteelpark
Arenberg 40, 3001 Leuven, Belgium