Search

Abstract Surrogate models play a vital role in overcoming the computational challenge in designing and analyzing nonlinear dynamic systems, especially in the presence of uncertainty. This paper presents a comparative study of different surrogate modeling techniques for nonlinear dynamic systems. Four surrogate modeling methods, namely, Gaussian process (GP) regression, a long short-term memory (LSTM) network, a convolutional neural network (CNN) with LSTM (CNN-LSTM), and a CNN with bidirectional LSTM (CNN-BLSTM), are studied and compared. All these model types can predict the future behavior of dynamic systems over long periods based on training data from relatively short periods. The multi-dimensional inputs of surrogate models are organized in a nonlinear autoregressive exogenous model (NARX) scheme to enable recursive prediction over long periods, where current predictions replace inputs from the previous time window. Three numerical examples, including one mathematical example and two nonlinear engineering analysis models, are used to compare the performance of the four surrogate modeling techniques. The results show that the GP-NARX surrogate model tends to have more stable performance than the other three deep learning (DL)-based methods for the three particular examples studied. The tuning effort of GP-NARX is also much lower than its deep learning-based counterparts.

Existing tools to detect and visualize severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) suffer from low selectivity, poor cell permeability, and high cytotoxicity. Here we report a novel self-immolative fluorescent probe (MP590) for the highly selective and sensitive detection of the SARS-CoV-2 main protease (Mpro). This fluorescent probe was prepared by connecting a Mpro-cleavable peptide (N-acetyl-Abu-Tle-Leu-Gln) with a fluorophore (i.e., resorufin) via a self-immolative aromatic linker. Fluorescent titration results show that MP590 can detect Mpro with a limit of detection (LoD) of 35 nM and is selective over interferents such as hemoglobin, bovine serum albumin (BSA), thrombin, amylase, SARS-CoV-2 papain-like protease (PLpro), and trypsin. The cell imaging data indicate that this probe can report Mpro in HEK 293T cells transfected with a Mpro expression plasmid as well as in TMPRSS2-VeroE6 cells infected with SARS-CoV-2. Our results suggest that MP590 can both measure and monitor Mpro activity and quantitatively evaluate Mpro inhibition in infected cells, making it an important tool for diagnostic and therapeutic research on SARS-CoV-2.

Abstract: Physics-based digital twins often require many computations to diagnose current and predict future damage states in structures. This research proposes a novel iterative global–local method, where the local numerical model is replaced with a surrogate to simulate cracking quickly on large steel structures. The iterative global–local method bridges the scales from the operational level of a large steel structure to that of a cracked component. The linear global domain is efficiently simulated using static condensation, and the cracked local domain is quickly simulated using the adaptive surrogate modeling method proposed herein. This work compares solution time and accuracy of the proposed surrogate iterative global–local method with a reference model, a submodeling model, and an iterative global–local method with no surrogate model for the local domain. It is found that the surrogate iterative global–local method gives the fastest solution time with comparatively accurate results.

Corrosion can initiate cracking that leads to structural integrity reduction. Quantitative corrosion assessment is challenging, and the modeling of corrosion-induced crack initiation is essential for model-based corrosion reliability analysis of various structures. This paper proposes a probabilistic computational analysis framework for corrosion-to-crack transitions by integrating a phase-field model with machine learning and uncertainty quantification. An electro-chemo-mechanical phase-field model is modified to predict pitting corrosion evolution, in which stress is properly coupled into the electrode chemical potential. A crack initiation criterion based on morphology is proposed to quantify the pit-to-cracking transition. A spatiotemporal surrogate modeling method is developed to facilitate this, consisting of a Convolution Neural Network (CNN) to map corrosion morphology to latent spaces, and a Gaussian Process regression model with a nonlinear autoregressive exogenous model (NARX) architecture for prediction of corrosion dynamics in the latent space over time. It enables the real-time prediction of corrosion morphology and crack initiation behaviors (whether, when, and where the corrosion damage triggers the crack initiation), and thus makes it possible for probabilistic analysis, with uncertainty quantified. Examples at various stress and corrosion conditions are presented to demonstrate the proposed computational framework.

Abstract: Local scour is a growing cause of bridge failure in the United States and around the world. In the next century, the effects of climate changes will make more bridges susceptible to scour failure more than ever before. This study aims to harness the spatially continuous monitoring capabilities of ultrasonic time-domain reflectometry to detect a soil interface for the purposes of scour monitoring. In this study, a long, slender plate is coupled with two flexible piezoelectric devices that propagate Lamb waves along the length of the plate to form the scour sensor. The sensor was tested for sensitivity to external pressure using metal weights and was able to detect the position of the pressure up at a length of up to ∼20 feet. The sensor was tested under simulated scour conditions, being buried in sand at various depths. The results show that the Lamb wave scour sensor is capable of reliably detecting a soil interface at 1 ft intervals. The scour sensor was also able to detect uncompacted soil interfaces, which is important considering the issue of scour hole refill following an extreme event.

Many physics-based and surrogate models used in structural health monitoring are affected by different sources of uncertainty such as model approximations and simplified assumptions. Optimal structural health monitoring and prognostics are only possible with uncertainty quantification that leads to an informed course of action. In this article, a Bayesian neural network using variational inference is applied to learn a damage feature from a high-fidelity finite element model. Bayesian neural networks can learn from small and noisy data sets and are more robust to overfitting than artificial neural networks, which make it very suitable for applications such as structural health monitoring. Also, uncertainty estimates obtained from a trained Bayesian neural network model are used to build a cost-informed decision-making process. To demonstrate the applicability of Bayesian neural networks, an example of this approach applied to miter gates is presented. In this example, a degradation model based on real inspection data is used to simulate the damage evolution.

Analogous to an experiment, a structural health monitoring (SHM) system may be thought of as an information-gathering mechanism. Gathering the information that is representative of the structural state and correctly inferring its meaning helps engineers (decision-makers) mitigate possible losses by taking appropriate actions (risk-informed decision-making). However, the design, research, development, installation, maintenance, and operation of an SHM system are an expensive endeavor. Therefore, the decision to invest in new information is rationally justified if the reduction in the expected losses by utilizing newly acquired information is more than the intrinsic cost of the information acquiring mechanism incurred over the lifespan of the structure. This article investigates the economic advantage of installing an SHM system for inference of the structural state, risk, and lifecycle management by using the value of information (VoI) analysis. Among many possible choices of SHM system designs (different information-gathering mechanisms), pre-posterior decision analysis can be used to select the most feasible design. Traditionally, the cost–benefit analysis of an SHM system is carried out through pre-posterior decision analysis that helps one evaluate the benefit of an experiment or an information-gathering mechanism using the expected value of information metric. This study proposes an alternate normalized metric that evaluates the expected reward ratio (benefit/gain of using an SHM system) relative to the investment risk (cost of SHM over the lifecycle). The analysis of evaluating the relative benefit of various SHM system designs is carried out by considering the concept of the VoI, by performing pre-posterior analysis, and the idea of a perfect experiment is discussed.

Analogous to an experiment, a structural health monitoring (SHM) system may be thought of as an information-gathering mechanism. Gathering the information that is representative of the structural state and correctly inferring its meaning helps engineers (decision-makers) mitigate possible losses by taking appropriate actions (risk-informed decision-making). However, the design, research, development, installation, maintenance, and operation of an SHM system are an expensive endeavor. Therefore, the decision to invest in new information is rationally justified if the reduction in the expected losses by utilizing newly acquired information is more than the intrinsic cost of the information acquiring mechanism incurred over the lifespan of the structure. This article investigates the economic advantage of installing an SHM system for inference of the structural state, risk, and lifecycle management by using the value of information (VoI) analysis. Among many possible choices of SHM system designs (different information-gathering mechanisms), pre-posterior decision analysis can be used to select the most feasible design. Traditionally, the cost–benefit analysis of an SHM system is carried out through pre-posterior decision analysis that helps one evaluate the benefit of an experiment or an information-gathering mechanism using the expected value of information metric. This study proposes an alternate normalized metric that evaluates the expected reward ratio (benefit/gain of using an SHM system) relative to the investment risk (cost of SHM over the lifecycle). The analysis of evaluating the relative benefit of various SHM system designs is carried out by considering the concept of the VoI, by performing pre-posterior analysis, and the idea of a perfect experiment is discussed.