Jaesung Lee

Accurately estimating spatially heterogeneous elasticity parameters, particularly Young's modulus and Poisson's ratio, from noisy displacement measurements remains a significant challenge in inverse elasticity problems. Existing inverse estimation techniques are often limited by instability, high noise sensitivity, and difficulties in recovering the absolute scale of Young's modulus. This work presents a novel Inverse Elasticity Physics‐Informed Neural Network (IE‐PINN) to robustly reconstruct heterogeneous elasticity distributions from noisy displacement data based on the principles of linear elasticity. The IE‐PINN incorporates three distinct neural network architectures, each dedicated to modeling displacement fields, strain fields, and elasticity distributions. This approach significantly enhances stability and accuracy under measurement noise. Additionally, a two‐phase estimation strategy is proposed: the first phase recovers relative spatial distributions of Young's modulus and Poisson's ratio, while the second phase calibrates the absolute scale of Young's modulus using boundary loading conditions. Methodological innovations, including positional encoding, sine activation functions, and a sequential pretraining strategy, further improve the model's performance and robustness. Extensive numerical experiments demonstrate that IE‐PINN effectively overcomes critical limitations faced by existing methods, providing accurate absolute‐scale elasticity estimations even under severe noise conditions. This advancement holds substantial potential for clinical imaging diagnostics and mechanical characterization, where measurements typically encounter substantial noise.

The rapid advancement of sensor technologies has made multi-sensory signals easily accessible in various manufacturing systems. To improve health condition assessment and prediction, a variety of neural network-based data fusion approaches have been proposed to generate health indices by fusing multi-sensory signals. However, most of the existing methods focus solely on using health indices to predict system failure times, overlooking their potential application in corresponding operational decision-making. Our focus is on this new usage of health indices to alert users to the warning stage of each system, thereby minimizing decision-making (maintenance) costs. To construct this novel type of health indices, this study proposes a systematic approach based on the concept of a ‘low-loss curve’ on the loss surface, enabling the training of a fusion neural network that provides health indices yielding both high remaining useful lifetime prediction accuracy and low maintenance costs. The effectiveness of the proposed method is demonstrated using a case study on aircraft turbofan engines, where it successfully reduces maintenance costs without sacrificing prediction accuracy. The code for the proposed methodology will be made publicly available on GitHub upon the publication of this paper.

Modern industrial systems are often subject to multiple failure modes, and their conditions are monitored by multiple sensors, generating multiple time-series signals. Additionally, time-to-failure data are commonly available. Accurately predicting a system's remaining useful life (RUL) requires effectively leveraging multi-sensor time-series data alongside multi-mode failure event data. In most existing models, failure modes and RUL prediction are performed independently, ignoring the inherent relationship between these two tasks. Some models integrate multiple failure modes and event prediction using black-box machine learning approaches, which lack statistical rigor and cannot characterize the inherent uncertainty in the model and data. This paper introduces a unified approach to jointly model the multi-sensor time-series data and failure time concerning multiple failure modes. This proposed model integrate a Cox proportional hazards model, a Convolved Multi-output Gaussian Process, and multinomial failure mode distributions in a hierarchical Bayesian framework with corresponding priors, enabling accurate prediction with robust uncertainty quantification. Posterior distributions are effectively obtained by Variational Bayes, and prediction is performed with Monte Carlo sampling. The advantages of the proposed model is validated through extensive numerical and case studies with jet-engine dataset.

Accurately estimating spatially heterogeneous elasticity parameters, particularly Young's modulus and Poisson's ratio, from noisy displacement measurements remains significantly challenging in inverse elasticity problems. Existing inverse estimation techniques are often limited by instability, pronounced sensitivity to measurement noise, and difficulty in recovering absolute-scale Young's modulus. This work presents a novel Inverse Elasticity Physics-Informed Neural Network (IE-PINN) specifically designed to robustly reconstruct heterogeneous distributions of elasticity parameters from noisy displacement data based on linear elasticity physics. IE-PINN integrates three distinct neural network architectures dedicated to separately modeling displacement fields, strain fields, and elasticity distributions, thereby significantly enhancing stability and accuracy against measurement noise. Additionally, a two-phase estimation strategy is introduced: the first phase recovers relative spatial distributions of Young's modulus and Poisson's ratio, and the second phase calibrates the absolute scale of Young's modulus using imposed loading boundary conditions. Additional methodological innovations, including positional encoding, sine activation functions, and a sequential pretraining protocol, further enhance the model's performance and robustness. Extensive numerical experiments demonstrate that IE-PINN effectively overcomes critical limitations encountered by existing methods, delivering accurate absolute-scale elasticity estimations even under severe noise conditions. This advancement holds substantial potential for clinical imaging diagnostics and mechanical characterization, where measurements typically encounter substantial noise.

Agriculturally derived biowastes can be transformed into a diverse range of materials, including powders, fibers, and filaments, which can be used in additive manufacturing methods. This review study reports a study that analyzes the existing literature on the development of novel materials from agriculturally derived biowastes for additive manufacturing methods. A review was conducted of 57 selected publications since 2016 covering various agriculturally derived biowastes, different additive manufacturing methods, and potential large-scale applications of additive manufacturing using these materials. Wood, fish, and algal cultivation wastes were also included in the broader category of agriculturally derived biowastes. Further research and development are required to optimize the use of agriculturally derived biowastes for additive manufacturing, particularly with regard to material innovation, improving print quality and mechanical properties, as well as exploring large-scale industrial applications.

In engineering system design, minimizing the variations of the quality measurements while guaranteeing their overall quality up to certain levels, namely the robust parameter design (RPD), is crucial. Recent works have dealt with the design of a system whose response-control variables relationship is a deterministic function with a complex shape and function evaluation is expensive. In this work, we propose a Bayesian optimization method for the RPD of stochastic functions. Dual stochastic response models are carefully designed for stochastic functions. The heterogeneous variance of the sample mean is addressed by the predictive mean of the log variance surrogate model in a two-step approach. We establish an acquisition function that favors exploration across the feasible and optimality-improvable regions to effectively and efficiently solve the stochastic constrained optimization problem. The performance of our proposed method is demonstrated by the extensive numerical and case studies. Note to Practitioners —Many manufacturing processes involve undesirable variations, which create variations in the final products. For example, many emerging manufacturing processes, such as nanomanufacturing, involve complex physical and chemical dynamics and transformation, creating variations in the manufacturing output. In such processes, it is crucial to design the manufacturing processes or products so that they have minimum variations in their quality. Meanwhile, it is also important to maintain the overall quality of the designed processes or products. Furthermore, acquiring data from many advanced manufacturing processes is often very costly, especially in the designing stage. In this work, we propose a data-driven method that automatically finds the best setting of manufacturing processes or products with the minimum variations of quality and a given constraint on the average quality satisfied. Our proposed method is used before conducting every experiment; It analyzes the historical data from previous experiments and provides a setting to be used in the next experiment. Our proposed method efficiently utilizes the historical data, and thus finds the best robust setting by conducting only a small number of experiments.

Structural health monitoring (SHM) is of significant importance in the operation of engineering systems to ensure the durability and reliability. In this article, we introduce a Bayesian optimization method using a multioutput Gaussian process to solve the structural fault diagnosis problem. This method utilizes a high fidelity finite element model (FE) of the structure and the impedance/admittance measurements from the structure to identify the location and severity of the damage. The method improves the accuracy of the damage diagnosis by adopting a multioutput Gaussian process as the surrogate model for the full FE model and Thompson sampling approach is used to guide the search for the structural damage in the Bayesian optimization. The detailed algorithms are presented, and the convergence analysis of the method is conducted. We apply our proposed method on simulated synthetic functions and it achieves better performance and higher convergence speed than the traditional mixed input optimization methods. We then apply our method on a real world structural damage identification problem using measured piezoelectric admittance data and illustrate the effectiveness of the proposed method.