Dinh Chi Pham

Joule-heated electric glass melters are highly energy-intensive and operate under stringent throughput, quality, and refractory-protection constraints, yet existing optimization approaches either rely on offline high-fidelity CFD—which is too slow for day-to-day operating decisions—or on purely data-driven energy models that do not enforce physics-based glass-quality constraints. This work presents a simulation-informed supervisory optimization framework that closes this gap by integrating (i) a high-fidelity CFD design of experiments on a cold-top electric glass melter, (ii) interpretable linear reduced-order models (ROMs) for furnace temperatures and glass-process indices (residence time, sand dissolution, mixing, and melting indices), and (iii) a convex linear-program supervisory optimizer that minimizes electrical power subject to throughput, glass-quality, and refractory-protection constraints. Sixteen CFD operating points spanning power inputs of 900–1671 kW, pull rates of 0.12–0.36 kg/s, and cullet fractions of 0–30% are used to fit and cross-validate the ROMs. The novelty of this contribution is the unification of CFD-derived multi-KPI quality constraints with a fast, interpretable supervisory optimizer suitable for advisory deployment on industrial electric melters. For a representative 1350 kW baseline, the optimizer recommends a reduced operating power of 1256 kW (~7% reduction) while satisfying all quality, throughput, and refractory constraints—demonstrating a practical ROM-enabled digital-twin pathway for energy-efficient electric glass melting.

Interfaces in polymer-based carbon fiber composites are critical regions that are most susceptible to delamination under static and impact loads. The interface strength and fracture toughness can be improved significantly by structural micro-reinforcements using polymer additive manufacturing (PAM). A micromechanics embedded macro-model is developed for the evaluation and extraction of interface properties of laminated coupons with printed polymer reinforcements (PPRs) at ply interfaces to efficiently tailor the interface properties. Given the microstructure of the printed reinforcements, an automatic cohesive insertion module is developed to simulate the improved interface fracture toughness resulting from a wavy crack path steering around PPRs. The applicability and accuracy of the developed modeling approach are demonstrated using ENF coupons with PPRs fabricated and tested. The validated modeling approach is then applied to explore the effect of geometric configurations of PPRs through double layer printing on the enhanced Mode II energy release rate. The developed numerical model provides a digital testing tool for tailoring interface properties, optimization of PPRs, and extraction of enhanced interface properties for delamination failure prediction of a large scale composite structure.

Defects are inevitable in the manufacture and service of composite laminates and their influence on the mechanical performance of composite laminates can be substantial. Ply waviness and delamination are two major types of defects. Compared to the tensile response, ply waviness has more significant influence on the compressive strength of composite structures. To evaluate the effects of ply waviness, experimental and numerical studies are performed. Ply waviness is generated by embedding composite strips in the fabrication of composite laminates. Photomicrographs are adopted to measure the ply waviness. Ply waviness is incorporated into the numerical model using the defects characterization and damage mapping method in our Composite Bolted and Bonded Analysis Toolkit for Abaqus (CB2ATA). A continuum damage mechanics (CDM) approach based on LaRC05 was developed and it is implemented in Abaqus with an implicit scheme. The CDM approach is then adopted in the numerical simulation of composite laminates under compression and the predicted compressive responses are compared with the experimental measurements.

Hole drilling is one of the most widely used operations in order to place bolts or rivets. Drilling holes in composites leads to drilling-induced defects and damages such as delamination, burr, microcracking, swelling, splintering, and fiber pullout. A reliable and effective simulation toolkit for composite drilling process is essential to capture the drilling induced damage and perform optimization of drilling parameters to achieve desired hole quality. Under this study, a combined experimental and numerical approach is developed to extract the composite response and damage distribution caused by the drilling operation for model exploration and validation. With an advanced reaction force evaluation system and the use of non-destructive inspection techniques, specimens of different configurations are drilled and postdrilling damage quantification are evaluated for validation of the toolkit. An enhanced user-defined material model is developed for the ABAQUS/explicit to simulate the time-dependent material removal process during drilling and associated damage distribution. Accounting for the effect of the shear dominant cutting during the drilling, enhanced damage models are implemented based on the use of a fiberorientation dependent shear failure criterion coupled with a failure mechanism driven energy dissipation for the post-peak material characterization. The applicability and high-fidelity of the developed toolkit are demonstrated by comparing both the global response and local failure predictions with the testing results of quasi-isotropic coupons with a backing plate of two different configurations.

Defects Characterization and Evaluation of Their Effects on Compressive Failure of Composite Components

Life extension is an attractive option for subsea flexible risers nearing the end of their design lives. However, techniques for assessing accumulated fatigue damage in flexible risers are often associated with large uncertainties due to the simplified calculation approaches typically used. One approach to reducing uncertainties is the inclusion of nonlinearities in riser structural response and consistent linking between global and local models. In this article, we present the elements of a numerical multiscale procedure capable of predicting the stresses that lead to fatigue damage in flexible pipes, namely: a nonlinear beam element, a nonlinear section response model and a detailed finite element model; the consistent integration of models developed for different length scales; and finally a validation of the flexible riser large-scale model.

In this paper, an enhanced three-dimensional continuum damage mechanics model is applied to predict the drop-weight impact response and compression after impact failure of a fiber-reinforced polymer composite specimen. The three-dimensional progressive damage model incorporates a three-dimensional maximum stress criterion to predict the intra-ply damage initiation, followed by a fracture-energy-based smeared crack model to capture the post-peak softening behavior. Driven by the dominant through-the-thickness failure under impact loading, a three-dimensional continuum damage model is implemented for the three-dimensional solid element via its explicit material model for Abaqus (VUMAT) to capture the effect of three-dimensional stress state and the interaction of matrix cracking and delamination. Abaqus’ restart analysis capability is used to activate the compression after impact analysis using the final damage state from the dynamic impact analysis. Both the dynamic failure and the compression after impact are demonstrated via a suite of verification examples followed by the sensitivity analysis using distinct impact configurations. The predictive capability of the proposed three-dimensional damage model is first verified using a static open-hole tension test. Applications of the damage model are then demonstrated for simulations of the dynamic drop-weight tests and compression after impact tests. A comparative study on the developed method is performed using the results predicted from the open-source CompDam. A sensitivity study is also performed to demonstrate the impact energy-dependent failure mode. The proposed model has shown its advantages in performing a quick assessment of impact damage and its effects on the residual compressive strength.

The use of composite materials has constantly increased in aerospace and offshore applications. One of typical composite structures for offshore application is multilayer unbonded composite pipe. The composite pipe comprised of several layers such as helically wound steel layers and polymeric layers, among which the helically wound steel wires play essential roles in providing axial and flexural stiffnesses of the pipe. Because of the geometric nonlinearity of each layer of the composite pipe and the flexibility in motion of individual layers against their adjacent layers, analyses of the composite pipe are highly nonlinear and computationally challenging. In this study, detailed three-dimensional Representative Volume Element models of the multilayer unbonded composite pipe are developed, and nonlinear analyses of the composite pipe are performed. The developed FE models effectively predict nonlinear responses of the pipe including slipping effects of steel layers under application of axial loading, bending, internal and external pressures. Based on the predicted results, a constitutive model of the pipe under various loadings is obtained which can be employed in the large scale analysis of the composite pipe to improve the accurate and computational efficiency of the global analysis.