Search

The existing machine learning (ML) models for vibration-based damage assessments often use highly compressed features or are restricted to preprocessed signals with fixed durations and sampling rates. Additionally, the learning capacities of ML models and computational resources are limited, which restricts using raw signals as direct input. This paper studied Mel filter banks (MFBs) for seismic signal processing, inspired by speech recognition technology. It is argued that the same filter designs in audio engineering may not be appropriate for seismic records, and therefore, a customized filter bank formulation was developed. Hybrid deep learning models for rapid assessments (HyDRA) were introduced as multibranch neural network architectures that enable end-to-end training for different types of processed vibration data structures. Moreover, the performance-based earthquake engineering (PBEE) equation was adjusted to integrate ML model uncertainties for probabilistic assessments. The proposed concepts were validated in a case study based on a data set of 32,400 nonlinear time-history analyses of a highway bridge in California. Insights and guidelines are provided for optimum filter design based on 5,184 experiments. Several HyDRA architectures were compared with benchmark models. The optimized MFB feature type outperformed features obtained from continuous wavelet transform and a stacked vector of conventional earthquake engineering indexes. A Bayesian variant of HyDRA was investigated to showcase its integration in the modified PBEE equation. Adopting custom filter banks with the HyDRA architecture enables effective feature extraction from raw vibration records by diversifying feature space and preserving information in the time and frequency domains. [ABSTRACT FROM AUTHOR]

Optimal sensor placement (OSP) is essential for effective structural health monitoring (SHM). More recently, deep learning algorithms have shown great potential in sensor‐based SHM. However, existing optimization frameworks, such as population‐based algorithms, are often not suited for data‐driven SHM. Evaluating a number of sensor layouts includes training on large datasets, which is computationally expensive. This paper proposes deep generative Bayesian optimization (DGBO) as a solution for a parallel optimization of black‐box/expensive OSP objective functions. Conditional variational autoencoders are leveraged as generative models that transform the OSP problem into a lower‐dimensional latent space. Additionally, DGBO utilizes a surrogate neural network to capture the probability distribution of the objective function space. The proposed method is validated on two case studies on a nine‐story reinforced concrete moment frame. The first one serves as a proof of concept to show that DGBO can find the global optimum configuration. The second case study aims to maximize the semantic damage segmentation (SDS) accuracy using a fully convolutional neural network. Transfer learning is proposed in training the vibration‐based SDS model, which reduces the evaluation times by more than 50%. Without compromising the performance, the number of accelerometers can be reduced by 52% and 43%, respectively, for damage location and severity predictions. It is also shown that DGBO can outperform genetic algorithm with the same number of function evaluations. DGBO can serve as a scalable solution to address the high‐dimensionality challenge in OSP for large‐scale civil infrastructure. [ABSTRACT FROM AUTHOR]

Automation in structural health monitoring (SHM) has greatly benefited from computer science's recent advances. Unlike images, the existing datasets for other types of input, such as vibration‐based damage data, are relatively smaller, less diverse, and highly imbalanced. Therefore, the reliability of data‐driven models developed for safety‐critical assessments can be questionable. This paper proposes a dual Bayesian inference where damage predictions are accompanied by measuring the model's confidence in predictions. First, it is shown how dual classification is integrated with Bayesian inference. Later, we introduce a surrogate deep learning module to transform the raw uncertainty output into an easily interpretable prediction uncertainty index (PUI). The PUI metric can be used to alarm a decision‐maker of the potential mistakes. The proposed dual Bayesian models are investigated on a 2D structure with seven different sensor layouts. Our approach yields increased robustness for different metrics compared with the benchmark. In addition to the performance boost, PUI information paves the way for a risk‐informed implementation of deep learning models in vibration‐based damage diagnosis. [ABSTRACT FROM AUTHOR]