Saurabh Prabhu

Feature Assimilation for Vibration Based Damage Detection

Structural health monitoring (SHM) technology for the early detection and mitigation of adverse structural effects, such as degradation or damage, is useful for enhancing the proactive maintenance of civil infrastructure. SHM techniques are advantageous because they eliminate the need for both a priori knowledge of the location of damage and access to the damaged portion of the structure. The underlying principle behind SHM involves measuring changes in a system’s vibration response, which ultimately indicate changes in physical properties due to structural damage. A challenge to the successful application of SHM to civil infrastructure is the selection of suitable vibration response features that are highly sensitive to the presence and extent of damage while also having low sensitivity to extraneous noise. This study reveals that both damage and the noise sensitivity of vibration response features vary for different states of structural health; therefore, the selection of optimum features is dependent on the damage severity, which is of course not known a priori. This study illustrates that assimilating multiple low-dimensional features lessens this dependence and improves the sensitivity of the damage indicators for SHM diagnosis.

Selection of Optimal Sensor Locations Based on Modified Effective Independence Method: Case Study on a Gothic Revival Cathedral

AbstractThis study demonstrates a methodology for selecting optimal regions for placing vibration sensors to efficiently extract the dynamic characteristics of the nave of the Cathedral of St. John the Divine (SJD). Placing sensors at optimal regions reduces not only the cost and time demands of vibration testing but also the amount of measurement data that needs to be postprocessed. Optimal sensor locations are determined by exploiting the finite-element (FE) model of the cathedral nave, built according to available documentation and best-engineering judgment and correlated against on-site measurement and inspection data. Optimal sensor locations are determined through a modified version of the effective independence method (MEIM) using the mode shape predictions of the correlated FE model. In (MEIM), there are two conflicting objectives: maximizing the orthogonality of mode-shape vectors of interest through an information-gain-based criterion while effectively exploring the geometry of the structure and...

Uncertainty Quantification Metrics with Varying Statistical Information in Model Calibration and Validation

Test-analysis comparison metrics are mathematical functions that provide a quantitative measure of the agreement (or lack thereof) between numerical predictions and experimental measurements. While...

Adaptively Weighted Support Vector Regression: Prognostic Application to a Historic Masonry Fort

AbstractPrognostic evaluation involves constructing a prediction model based on available measurements to forecast the health state of an engineering system. One particular prognostic technique, support vector regression, has had successful applications because of its ability to compromise between fitting accuracy and model complexity in training prediction models. In civil engineering applications, compromise between fitting accuracy and model complexity depends primarily on the measured response of the system to loads other than those that are of interest for prognostic evaluation, referred to as extraneous noise in this paper. To achieve accurate prognostic evaluation in the presence of such extraneous noise, this paper presents an approach for optimally weighing fitting accuracy and complexity of a support vector regression model in an iterative manner as new measurements become available. The proposed approach is demonstrated in prognostic evaluation of the structural condition of a historic masonry ...

Assessment of Strength Degradation of Historic Masonry Monuments due to Damage: Load Path–Based Approach

AbstractHistoric masonry monuments experience aging and accumulated damage, which consequently degrade both their stiffness and strength. Stiffness degradation can be quantified through nondestruct...

Model assessment in scientific computing: Considering robustness to uncertainty in input parameters

Purpose This paper aims to focus on the assessment of the ability of computer models with imperfect functional forms and uncertain input parameters to represent reality. Design/methodology/approach In this assessment, both the agreement between a model’s predictions and available experiments and the robustness of this agreement to uncertainty have been evaluated. The concept of satisfying boundaries to represent input parameter sets that yield model predictions with acceptable fidelity to observed experiments has been introduced. Findings Satisfying boundaries provide several useful indicators for model assessment, and when calculated for varying fidelity thresholds and input parameter uncertainties, reveal the trade-off between the robustness to uncertainty in model parameters, the threshold for satisfactory fidelity and the probability of satisfying the given fidelity threshold. Using a controlled case-study example, important modeling decisions such as acceptable level of uncertainty, fidelity requirements and resource allocation for additional experiments are shown. Originality/value Traditional methods of model assessment are solely based on fidelity to experiments, leading to a single parameter set that is considered fidelity-optimal, which essentially represents the values which yield the optimal compensation between various sources of errors and uncertainties. Rather than maximizing fidelity, this study advocates for basing model assessment on the model’s ability to satisfy a required fidelity (or error tolerance). Evaluating the trade-off between error tolerance, parameter uncertainty and probability of satisfying this predefined error threshold provides us with a powerful tool for model assessment and resource allocation.

Model Validation in Scientific Computing: Considering Robustness to Non-probabilistic Uncertainty in the Input Parameters

The origin of the term validation traces to the Latin valere, meaning worth. In the context of scientific computing, validation aims to determine the worthiness of a model in regard to its support of critical decision making. This determination of worthiness must occur in the face of unavoidable idealizations in the mathematical representation of the phenomena the model is intended to represent. These models are often parameterized further complicating the validation problem due to the need to determine appropriate parameter values for the imperfect mathematical representations. The determination of worthiness then becomes assessing whether an unavoidably imperfect mathematical model, subjected to poorly known input parameters, can predict sufficiently well to serve its intended purpose. To achieve this, we herein evaluate the agreement between a model’s predictions and associated experiments as well as the robustness of this agreement given imperfections in both the model’s mathematical representation of reality as well as its input parameter values.

Structural Assessment of Fort Sumter Masonry Coastal Fortification Subject to Foundation Settlements

In historic unreinforced masonry structures, one of the most critical structural issues is the differential settlement of foundations. Due to unreinforced masonry’s brittle nature with a fairly low tensile strength, brittle cracking can occur due to tensile stresses introduced by foundation settlements. This study demonstrates the development and calibration of a finite element model and the use of this model for structural analysis under differential settlements of a casemate of Fort Sumter, a masonry coastal fortification best known as the site where the first shots of The American Civil War were fired in 1861. Development of accurate finite element models for historic masonry structures presents numerous challenges in the acquisition of non-linear material properties, and irregular geometry. Furthermore, these challenges are exacerbated because of the configuration of coastal fortifications, as these structures have characteristic designs unique to the distinct functionality of defense, such as cold-joints between disjointed structural components. The non-linear material behavior of the brick and mortar assembly is obtained from laboratory tests on samples obtained at the site. High definition laser scanning is used on irregular geometry to obtain the details of accumulated structural damage and degradation, including differential foundation settlements. The uncertain interface behavior at the cold joint between the scarp wall and the casemate of the fort is assessed using in-situ vibration tests. The finite element model developed is utilized to study the settlement magnitudes critical to the stability of the casemates of Fort Sumter for a variety of possible soil settlement configurations.

Feature Assimilation in Structural Health Monitoring Applications

Next generation structural health monitoring (SHM) technology for early detection and mitigation of adverse structural effects holds the potential to aid in the proactive maintenance of various civil structures. SHM techniques eliminate the need for a priori knowledge of damage, and thus the need for access to the damaged portion of the structure. The underlying principle behind SHM is measuring changes in the system vibration response, which would ultimately indicate changes in the physical properties due to structural damage. A challenge to the successful application of SHM to civil structures is the selection of suitable vibration response features (damage indicators), that are highly sensitive to the presence and extent of damage, while having low sensitivity to ambient noise. Since it is not feasible (nor possible) to damage an in-service structure for research purposes, a scaled arch model made of PVC is utilized for laboratory testing in this study. The vibration response is measured both for the undamaged arch and then for the damaged arch once cracks are introduced to the system. The effect of noise on the vibration measurements is also studied.