See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/270450937
Structural Health Monitoring Standards Conference Paper · September 2014 DOI: 10.2749/222137814814069804
CITATIONS
READS
0
381
1 author: Andrea Enrico Del Grosso Università degli Studi di Genova 58 PUBLICATIONS 181 CITATIONS SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Structural Health Monitoring, lifecycle and BIM View project
All content following this page was uploaded by Andrea Enrico Del Grosso on 16 June 2015. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.
Structural Health Monitoring Standards Andrea E. DEL GROSSO Professor of Engineering University of Genoa Genoa, Italy
[email protected]
Andrea E. Del Grosso, born in 1945, graduated in Civil Engineering at the University of Genoa, Italy, where he became full Professor of Structural Engineering. His research interests include seismic engineering, innovative structural systems, smart structures and structural health monitoring.
Summary Although Structural Health Monitoring (SHM) techniques can be considered relatively mature, at least from the scientific point of view, they have not yet become a standard practice in civil engineering because of several reasons. One of these reasons has been individuated in the lack of comprehensive standards addressing the complete SHM process and especially potential utilization. The paper is aimed at briefly reviewing existing standards and tracing the lines for possible extensions, based on experiences gathered from the field and the actual evolution of SHM techniques. Lifecycle approaches including SHM are considered as well. In addition, the relationships between Structural Health Monitoring and Design Standards are addressed, raising the question whether in the design of a monitored structure and of a traditional one, the same safety coefficient should be applied. Consideration is given to the aspect of uncertainty modelling in both cases. Keywords: Structural Health Monitoring, Lifecycle Engineering, Structural Reliability, Standards.
1.
Introduction
In the last two decades, Structural Health Monitoring (SHM) techniques have received considerable attention from academics and engineering practitioners. A very significant number of dedicated conference series and journals have been established on the subject and many important societies, like IABSE, include SHM as an important topic in their general conferences. From the point of view of practical applications, the use of automated systems for integrity monitoring of assemblies and components is already common in automotive and aerospace engineering but, despite of the theoretical developments available to date and of the several successful application examples, civil engineers look to be somehow reluctant to extensively apply SHM techniques. Among the various reasons that may stay behind this attitude, the following are mentioned here. - Extreme diversity of the players involved in the civil engineering sector, including facility owners, designers and contractors, in terms of economic interests, size and technical skills. - Extreme diversity of the structural typologies and situations involved in the construction industry. - Lack of understanding of the potential benefits induced by the use of SHM techniques. - Extreme diversity of the players in the SHM market. Some aspects of the above reasons will be discussed in more detail in the present paper, putting into evidence the different roles that SHM could indeed play in improving knowledge on the real structural behaviour, in allowing effective management of constructed facilities and, generally speaking, in pushing structural engineering towards more effective, safe and sustainable practices.
Difficulties and unsolved problems shall not however be hidden. Reliability issues in assessing the structural conditions (damage states) from SHM data in real environments generate many open problems both from the theoretical and from the practical points of view. Durability, maintainability and lifecycle costs of SHM systems compared to the life expectancy of civil structures also pose important questions. Complexities and uncertainties in materials degradation and infrastructure obsolescence phenomena are important issues as well, still to be completely investigated in a SHM context. Nonetheless, it is a common belief among SHM experts that the development and implementation of Structural Health Monitoring standards and guidelines will be of a paramount importance in disseminating SHM practices and in helping to gather from a wider experience the information and knowledge necessary to bring open problems to solution. For this reason, this paper is dedicated to review and discuss some of the initiatives that have been recently started to develop such standards and guidelines. The discussion will be preceded by a presentation of the role that SHM can play in a lifecycle approach to structural engineering. The presentation will help the determination of the topics that should be covered by standards. The formulation will also be reconsidered at the end of the paper to discuss the relationships between SHM and design procedures, in the light of a comparison between the design safety factors and the actual safety factors when structural health monitoring techniques are systematically applied.
2. SHM in a lifecycle approach As anticipated in the introduction, advantages in using SHM in the civil structural engineering field are potentially very broad, ranging from providing a more comprehensive knowledge on the real structural behaviour and on the real loads, to validating design hypotheses and calculations, to disclose and quantify phenomena actually taking place and to serve as a basic tool for ensuring structural safety during service life, optimizing maintenance strategies and costs. The role of SHM with respect to this latter aspect will be specifically addressed in the framework of a lifecycle approach. Lifecycle approaches to design, construction and maintenance of buildings and infrastructure have been recently developed and their application proposed into practice. Safety and usability of existing structures actually depend on the level of degradation and obsolescence due to ageing; this consideration has been often referred to as the “time-dependent reliability problem”. The time-dependent reliability problem was first discussed with reference to ageing concrete structures by several authors, among which Mori &Ellingwood [1], Stewart &Rosowsky [2] and Enright &Frangopol [3]. In view of the application of the above concepts to infrastructure management, several reliabilitybased maintenance strategies, able to keep a global measure of safety and effectiveness above acceptable limits, have been subsequently proposed in terms of lifetime functions and life-cyclecost optimization [4,5,6]. Applications to bridges and to general structures form the subject of series of conferences organized by international associations like IABMAS [7] and IALCCE [8]. A lifetime function is a function expressing the decay of a performance parameter with time, for example the reliability index β or any other convenient measure of the safety or efficiency of a critical structural component or of the structure as a whole. The determination of a lifetime function from theoretical degradation models is very complex and severely affected by uncertainties so that, in practical applications, empirical mathematical expressions are preferred. Uncertainties in the knowledge and actual development of the phenomena represented by the lifetime function are both stochastic and epistemic in nature. The epistemic component will also make estimates of the probability distributions to widen with time and, since the phenomena shall be forecasted for the whole life of the structure, the trend of mean and confidence interval curves will be as qualitatively depicted in Figure 1. A proposal to standardize the lifetime functions and the corresponding probability distributions, based on statistical data collected on bridges, has resulted from the work recently performed by the CEN WG 63 [9]. The standardized function is expressed by a two-parameters exponential and a standard probability distribution is also suggested. The exponential formula is constructed in such a way to reach the acceptable limit of performance (failure) at the end of the design life and
may be used as an initial estimate of the ageing phenomenon or to evaluate the remaining life, for any given value of the condition index resulting from a condition determination process. Since a maintenance intervention has the effect of improving the condition index, the standardized function may be effectively used to the purpose of optimally planning maintenance strategies for ageing structures. When periodic assessments of the condition index Lf are performed by means of visual inspection programs or when a permanent structural health monitoring system is installed on the structure, the actual trend of the lifetime function may be updated and the associated uncertainties may be reduced. mean In particular, if the monitoring system is operated continuously, the sensors are producing a very large set of data that need to be organized, filtered, t validated and interpreted in order to produce a measure of the condition index suitable for the Fig. 1 – Typical lifetime function updating of the lifetime functions and of the related uncertainties. Of course the process itself is affected by the uncertainties associated with the sensor readings and data interpretation and the lifetime function updating should be defined accordingly, for example by means of a Bayesian scheme. The typical flow-chart of the lifetime functions updating process from SHM data is schematically represented in Figure 2 [10].
Sensor 1
Data PreProcessing
Damage Identification Process
Sensor 2
Data Fusion
Lifetime Function Update
Sensor n
Data PreProcessing Sensor m
Fig. 2 – Lifetime performance assessment flow-chart From the above discussion, it is possible to extract the characteristics of a structural health monitoring system and of its utilization process in the civil engineering field that should be addressed by standards and guidelines. These characteristics, or critical features, are presented in more detail in the next paragraph.
3. Critical features of SHM approaches A structural health monitoring system and its utilization process are composed by several subsystems and sub-processes, and each of them is intended to perform a different function at the hardware, software or at the conceptual level.
Typical subsystems, functions and sub-processes may be listed as follows: - the set of physical parameters governing the structural conditions and the condition index, - the instrumentation system that is used to measure the parameters, - the data logging and storage subsystem, - the signal (or data) processing tools that are used to validate the measurements and perform the data fusion function, - the damage identification algorithms, - the condition determination process and related models, - the process that is used to update the lifetime functions, - the process that is used to optimize maintenance strategies. In view of a systematic use of the SHM approach , the following aspects can be considered critical: - the significance of the physical parameters to be monitored, - the reliability, availability and serviceability of the sensory systems and other hardware components for the total duration of the expected life of the structure, - the reliability of the algorithms that are used to perform the damage identification function, in terms of likelihoods associated with the detection, localization and determination of the severity of damages or degradation. In addition, the significance of the condition index, the reliability of the models used for its estimate and of the subsequent processing steps should also be considered critical. These latter aspects, however, can be viewed to be somehow independent on the method adopted to construct the damage pattern present on the structure at a given time. Indeed, referring to procedures for the management of bridges and other constructed facilities based on visual inspection and conventional non-destructive testing, it can be recognized that experimented standards are already existing in many countries. To mention a few of them: AASHTO [11], DIN 1076, or the Austrian RVS 13-03-11 [12]. An interesting issue is eventually how lifetime functions based approaches relate to these conventional approaches in the maintenance optimization processes. This issue is however outside the scope of the present discussion. Coming back to SHM, in order to render effective the corresponding procedures in integrating and, in perspective, partially substituting conventional visual inspection approaches, standardization should address the critical aspects mentioned above.
4. SHM Standards and Guidelines Restricting the discussion to documents specifically addressing civil engineering structures, some guidelines for Structural Health Monitoring have been already published in the past. One of the first publications on the subject has been released by ISIS Canada [13]. This document is quite comprehensive, mentions several important aspects and introduces the new sensing technologies emerging at the time of publication. Basically, static and dynamic field testing, periodic and continuous static and dynamic monitoring are addressed, and the approach presented shows a conceptual continuity between well-established methods of testing real structures in the field and the new approach of Structural Health Monitoring. A classification and review of the important SHM system components is given in the text, together with some indication of the data processing needs for the treatment of monitoring data. The Manual, also presenting commented examples, has served as a practical reference for a long time. Other similar publications have however appeared more recently, like the document produced in the framework of the European research network SAMCO [14]. An interesting case is also represented by the Russian GOST R 53778-2010 [15]. In this
document, the use of which is mandatory in the Russian Federation, besides describing visual inspection and testing methods and condition-based classification schemes for different types of structures, identifies cases where SHM systems shall be necessarily applied, as well as the physical parameters that shall be taken under control. Following to the publication of a book by Wenzel on the SHM of bridges [16], that also contains useful guidelines for the conceptual design of SHM systems and for subsequent processing, mostly based on dynamic measurements, an official guideline for the monitoring of bridges and other engineering structures has been published in Austria in 2012 [17]. A guideline for the SHM of long-span bridges is being released in China [18]. ISO standards for measuring and processing the vibratory response of bridges and buildings should also be recalled [19-21]. To issue other reference documents coping with the new experiences and monitoring approaches, a number of initiatives have been very recently started in some countries. In Italy for example, UNI has established a Working Group to prepare a Structural Health Monitoring Guideline, to be essentially applied to buildings and bridges. Harmonization of standards at the European and International levels is also under way. An international initiative has indeed been launched during the 9th International Workshop on Structural Health Monitoring, held at Stanford University in September 2013. Looking to the content of the documents in more detail, in order to be effective in encouraging the use of SHM technologies, a new generation of guidelines and standards should be extended to explicitly consider, besides traditional subjects, the durability of SHM systems with respect to the expected operational life of the structures they can be permanently applied to. In particular, lifecycle, maintenance and procedures for substitution of the sensory systems shall be treated. Guidelines should also be opened to innovative sensory systems and smart materials. A very large number of different signal processing and damage identification (DI) algorithms have been proposed in the literature, but up to now only a few of them have been successfully applied in the field. Guidelines should therefore address the criteria for cleaning data from undesired environmental effects and for selecting the DI algorithms most appropriate to the case. They should consider the relationship among algorithms, location of sensors, performance parameters and phenomena to be observed, including the monitoring of loads. Procedures for evaluating the likelihood of the selected approaches, for example in terms of the Probability of Detection (POD) and Probability of False Alarm (PFA), and for updating structural models should be considered as well. Finally, the relationship between SHM-based lifetime functions and conventional facility management procedures based on visual inspections, NDE and condition-based classification should be clarified.
5. SHM, inspection and design standards As previously discussed, the development of a new generation of guidelines and standards for Structural Health Monitoring will promote the diffusion of practical applications. However, considering the full range of benefits that the diffusion of SHM techniques in the construction sector could provide, it is necessary to consider the perspective of reduction in the life-cycle cost. In recent studies performed by Frangopol and Liu [22], a comparison between optimal design solutions with and without monitoring has been already presented. The conclusions of the study are summarized in Figure 3. Although monitoring does not change the relationship between the costs (i.e. that higher initial costs result in lower failure and additional costs) monitoring does change each cost itself. Specifically, the initial cost is increased (upfront SHM system cost), the failure cost is decreased (less risk), and the additional costs are decreased (improved optimal management decisions) across the entire profile. This reduction in total life-cycle cost is also expected to be paired with a higher level of performance. To clarify the concept, the two following aspects will be considered in more detail: the cost of inspection, as for example required by current regulations on bridges (additional costs) and the construction cost, as deriving from the application of current design standards (initial cost). For both aspects, the assessment of the likelihood of SHM procedures is an essential issue.
Experience has demonstrated that, once a monitoring system has been installed and maintained in operation, the cost of a condition assessment based on SHM data is less than the corresponding assessment based on the combination of visual inspection and non-destructive testing. SHM derived information combined with model updating and optimization of maintenance strategies allow reduction of additional costs. However, conductance of visual inspection and NDT cannot be avoided. If the SHM approach can be proven to be as reliable as conventional processes (as a matter of fact, it is considered to be potentially more reliable), the frequency of conventional investigations could be Fig. 3 - Optimum design solution based on substantially reduced. It is expected that life-cycle cost minimization with and without future regulations will explicitly consider the monitoring. Solid line: without monitoring; benefits deriving from the application of dotted line: with monitoring [22] SHM techniques. As concerning the second aspect, the qualitative considerations expressed by Figure 3 indicate that the optimum design solution implies both a saving in the total cost and an improvement in performance; this is obtained despite of the increase in the initial cost. As already noted, the initial cost is composed by the construction cost plus the cost of installation of the SHM system. The construction cost is obviously depending on the target performance index (for example the reliability index β) while the cost of the SHM system can be considered constant. In the semi-probabilistic limit state approach to structural design, the partial safety factors depend on the index β and also take into account various forms of epistemic uncertainties that may affect the safety of the structure as built. Deployment of SHM techniques will generally improve knowledge of the phenomena actually taking place on the real structure thus reducing uncertainties and disclosing, at any required time, the real level of risk associated to keeping in service that structure. It is emphasized that this is not only a consideration related to the actual structural conditions, that may be eventually restored to the design levels by maintenance works, but it is also related to the general framework of the capacity-demand problem. Then, the question arises if the reduction of the uncertainties (epistemic) deriving from the deployment of SHM technologies can be accounted for in design standards, reducing the safety factors applicable to monitored structures with respect to conventional ones. Figure 3 suggests a qualitative answer: if the level of performance corresponding to the optimum design solution for a conventional structure (without monitoring) leads to an already acceptable risk, the gain in performance (reliability index β) corresponding to the optimum design solution for the monitored structure could be spent in a reduction of the partial safety factors and, consequently, in a further economical benefit because of the reduction of the construction cost. Of course, translating this qualitative finding into quantitative assessments and design procedures is not immediate and requires specific studies, field experience and global consensus. In the Author’s knowledge, as of today, there are no published studies facing the problem with such a comprehensive view but it is believed that such studies are needed in the near future. It is, again, emphasized that accepted methods to evaluate the reliability of SHM systems and the likelihood of the information provided by the SHM process are a prerequisite to consistently perform those studies. It is however noted that some research efforts have already been devoted to the subject (for example [10]). A very simple example has been recently developed by the Author and his students. It will be illustrated in detail in a different paper. The main aspects of the study are briefly summarized as follows.
With reference to previous experimental monitoring studies on corrosion detection in reinforced concrete beam specimens [23], the characteristics of which are depicted in Figure 4, the initial diameter (16 mm) of the main re-bars has been exactly correlated to a uniformly distributed flexural load using the partial safety factors of EUROCODE 2 for ultimate limit states (for steel, γS = 1.15).
Fig. 4 – Corrosion test specimen and sensor position (Sn: fiber optic strain sensors, A: accelerometers) A mathematical corrosion law expressing the decay of diameter has been formulated based on literature data. Confidence interval curves have also been extrapolated assuming normal distributions of the variables and performing a reference Monte Carlo simulation with real environmental data. The lower degradation curve of the diameter leads to reach the limit state in the re-bars (γS = 1.0) after 87 years. The scatter of the degradation curves is of course increasing with time. It has been assumed to develop a SHM program to measure the re-bar diameter starting from the 20th year of operation and repeating the measurements every 10 years. Because the scope of the example was to verify the influence of SHM upon uncertainties, it has been assumed that the measurements were giving as a mean the theoretic mean value of the degradation curve and a statistical uncertainty, represented by a normal distribution, kept constant during time, has been set by heuristically estimating a standard deviation from the concept of Receiver Operating Characteristic (ROC) curve as described in [10], using data gathered during the experiment [23] and other SHM experiences on reinforced concrete beams. By repeatedly applying a Bayesian update, the uncertainty on the degradation curve has been reduced, leading to reach the limit state well beyond the originally expected life. By scaling the curve values to have the same probability of failure after 87 years, it has been found that the application of a SHM program could have allowed to reduce the design material safety factor to γS = 1.10. This is a very simple example, but it may represent a very first insight to a much more complex problem.
6. Conclusions Some of the major difficulties affecting the adoption of SHM technologies as a common engineering practice have been reviewed in the paper. The need of developing a new generation of guidelines and standards has been pointed out and the existing documents on the subject have been commented. The benefits from the adoption of SHM techniques with respect to current regulations and standards in infrastructure management and structural design have also been pointed out, showing how monitored structures could theoretically been designed according to less conservative safety factors. Assuming these considerations as final objectives, a new generation of SHM guidelines and standards could be developed in a top-down approach.
References [1] [2] [3]
MORI Y. and ELLINGWOOD B. R., “Reliability-based Service-Life Assessment of Aging Concrete Structures”, J. of Structural Engineering,Vol. 119, No. 5, 1993, pp.1600-1621. STEWART M. G. and ROSOWSKY D. V., “Time-dependent reliability of Deteriorating Reinforced Concrete Bridge Decks”, Structural Safety, Vol. 20, No. 1, 1998, pp. 91-109. ENRIGHT M. P. and FRANGOPOL D., “Probabilistic Analysis of Resistance Degradation of Reinforced Concrete Bridge Beams under Corrosion”, Engineering Structures, Vol. 20,
[4]
[5] [6] [7] [8] [9] [10]
[11] [12]
[13] [14] [15]
[16] [17]
[18] [19] [20] [21] [22]
[23]
View publication stats
No. 11, 1998, pp. 960-971. MIYAMOTO A., KAWAMURA K. and NAKAMURA H., “Development of a Bridge Management System for Existing Bridges”, Advances in Engineering Software, Vol. 32, 2001, pp. 821-833. FRANGOPOL D.M. and LIU M., Life-cycle cost and performance of civil structures, McGraw-Hill 2006 Yearbook of Science and Technology, McGraw-Hill, New York, 2006. AKTAN A.E., ELLINGWOOD B.R. and KEHOE B., “Performance-based engineering of constructed systems”, Journal of Structural Engineering, Vol.133, No. 3, 2007, pp. 311-323. IABMAS Conferences,http://www.iabmas.org IALCCE Conferences,http://www.ialcce.org CEN/WS 063, CWA16633:2013, Structural Condition Determination for Integrated Lifetime Assessment of Plants, Structures and Components. DEL GROSSO A. and LANATA F., “Uncertainties in damage identification and lifetime functions of ageing concrete structures”, in Structural Health Monitoring 2011, F.K. Chang Ed., DESTech Publication, Lancaster, 2011, pp. 1871-1878 AASHTO, Manual for Bridge Evaluation, Second Edition, 2011 ÖSTERREICHISCHES FORSCHUNGSGESELLSCHAFT, RVS 13-03-11, Quality Assurance for Structural Maintenance, Surveillance, Checking and Assessment of Bridges and Tunnels, Road Bridges (in German), 2011 ISIS CANADA, Guidelines for Structural Health Monitoring, Design Manual No. 2, September 2001 RÜCKER W., HILLE F. and ROHRMANN R., Guideline for Structural Health Monitoring, SAMCO Final Report F08b, 2006 RUSSIAN FEDERATION, National Standard of the Russian Federation, GOST R 537782010, Building and Structures, Technical Inspections and Monitoring Regulations (English translation), 2010 WENZEL H., Health Monitoring of Bridges, John Wiley & Sons, Chichester, 2009 ÖSTERREICHISCHES FORSCHUNGSGESELLSCHAFT, RVS 13-03-01, Quality Assurance for Structural Maintenance, Surveillance, Checking and Assessment of Bridges and Tunnels, Monitoring of Bridges and other Engineering Structures (in German), 2012 OU J., oral communication, 9th International Workshop on Structural Health Monitoring, Stanford University, September 10-12, 2013 ISO 14963:2003, Mechanical Vibration and Shock – Guidelines for dynamic tests and investigations on bridges and viaducts ISO 18649:2004, Mechanical Vibrations – Evaluation of measurement results from dynamic tests and investigations on bridges ISO 16587:2004, Mechanical Vibration and Shock – Performance parameters for condition monitoring of structures FRANGOPOL D. M. and LIU M., “Maintenance and Management of civil infrastructure based on condition, safety, optimization, and life-cycle cost”, Structure and Infrastructure Engineering, Vol. 3 No. 1,2007, pp. 29-4 DEL GROSSO A., LANATA F., PARDI L. and MERCALLI A., “Health Monitoring for Corrosion Detection in Reinforced Concrete Bridges”, Proc. 4th Int. Conf. on Bridge Maintenance, Safety and Management (IABMAS), Seoul, Korea. Koh and Frangopol eds., Taylor and Francis, 2008