An official website of the United States government
Abstract Continuous efforts are underway for the reduction of the structural weight of transit through the introduction of a multi-material metal-composites system. There are major challenges in joining dissimilar materials to result in optimum structural integrity. The conventional joining techniques have limitations in terms of preparation time, weight penalty resulting from adhesives, and uncertainty in joint integrity. Recently adoption of macro scale mechanical interlocking in the adhesive joining resulted in significant improvement of joint performance. This made mechanical interlocking gain an attention for hybrid joining. In this study, fastenerless method of mechanical interlocking based on Japanese wood joining craft is considered for joining carbon fiber-reinforced polyamide thermoplastic composite to aluminum. Different interlocking joining designs (IJDs) were developed. The joints were obtained by force-fitting the male into the female counterpart. Here the male and female segments joined at macro level with no joining integrity at the interface. Further, these joints were tested and evaluated for tensile strength. A finite element analysis (FEA) model is developed for stress analysis and studying failure mechanisms of the IJDs. It was observed that the geometry of IJD dictates the failure mode and material composition governs the maximum strength achieved by a particular IJD. Each IJD showed higher load capacity with metal as a female counterpart to the composite compared to other way round.
more »
« less
This content will become publicly available on August 1, 2026
Physics-Informed Machine Learning for Failure Mode Proportion Prediction in Composite Adhesive Joints
Abstract Adhesive bonding of composite materials has become increasingly crucial for advanced engineering applications, offering unique advantages for lightweight and high-performance designs. This study presents a novel framework, physics-informed failure mode proportion prediction (PIFMP) model, for predicting failure mode proportions in composite adhesive joints, addressing critical gaps in understanding mixed-mode failure behaviors. In contrast to conventional approaches that focus solely on force or stress prediction, this research integrates important parameters from multistage manufacturing processes (MMPs) and simulation data into a physics-informed machine learning (PIML) framework, enabling proactive failure prediction and design optimization. The proposed framework unifies data-driven machine learning models with features derived from finite element analysis (FEA), incorporating cohesive zone modeling (CZM) to capture the physical dynamics of adhesive behavior under lap shearing. By embedding FEA-based physics features into the machine learning process and leveraging a time-series transformer model to analyze the temporal progression of interfacial damage and separation, the framework ensures predictive accuracy and physics-informed consistency, enabling precise analysis of failure mechanisms. The empirical study validates the effectiveness and the reliability of the framework, demonstrating enhanced predictive performance through cross-validation. The work establishes a foundational approach for failure analysis and provides a robust basis for future advancements.
more »
« less
- Award ID(s):
- 2052714
- PAR ID:
- 10609869
- Publisher / Repository:
- American Society of Mechanical Engineers (ASME)
- Date Published:
- Journal Name:
- Journal of Manufacturing Science and Engineering
- Volume:
- 147
- Issue:
- 8
- ISSN:
- 1087-1357
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Context. Machine-learning methods for predicting solar flares typically employ physics-based features that have been carefully cho- sen by experts in order to capture the salient features of the photospheric magnetic fields of the Sun. Aims. Though the sophistication and complexity of these models have grown over time, there has been little evolution in the choice of feature sets, or any systematic study of whether the additional model complexity leads to higher predictive skill. Methods. This study compares the relative prediction performance of four different machine-learning based flare prediction models with increasing degrees of complexity. It evaluates three different feature sets as input to each model: a “traditional” physics-based feature set, a novel “shape-based” feature set derived from topological data analysis (TDA) of the solar magnetic field, and a com- bination of these two sets. A systematic hyperparameter tuning framework is employed in order to assure fair comparisons of the models across different feature sets. Finally, principal component analysis is used to study the effects of dimensionality reduction on these feature sets. Results. It is shown that simpler models with fewer free parameters perform better than the more complicated models on the canonical 24-h flare forecasting problem. In other words, more complex machine-learning architectures do not necessarily guarantee better prediction performance. In addition, it is found that shape-based feature sets contain just as much useful information as physics-based feature sets for the purpose of flare prediction, and that the dimension of these feature sets – particularly the shape-based one – can be greatly reduced without impacting predictive accuracy.more » « less
-
Abstract Predicting failure in solids has broad applications including earthquake prediction which remains an unattainable goal. However, recent machine learning work shows that laboratory earthquakes can be predicted using micro-failure events and temporal evolution of fault zone elastic properties. Remarkably, these results come from purely data-driven models trained with large datasets. Such data are equivalent to centuries of fault motion rendering application to tectonic faulting unclear. In addition, the underlying physics of such predictions is poorly understood. Here, we address scalability using a novel Physics-Informed Neural Network (PINN). Our model encodes fault physics in the deep learning loss function using time-lapse ultrasonic data. PINN models outperform data-driven models and significantly improve transfer learning for small training datasets and conditions outside those used in training. Our work suggests that PINN offers a promising path for machine learning-based failure prediction and, ultimately for improving our understanding of earthquake physics and prediction.more » « less
-
This work describes a method in which the digital image correlation (DIC) method and finite element analysis (FEA) were used to create a quasi-static mixed-mode fracture envelope for bonded joints consisting of 2024-T3 Al adherends and a tough structural epoxy adhesive. Symmetric and asymmetric versions of double cantilever beam, single-leg bend, and end-notched flexure tests are used to populate the mixed-mode fracture envelope with results at several mode mixities. Experiments are conducted in a universal testing machine while recording images for subsequent DIC analysis. Finite element analysis is used to implement cohesive zone models (CZMs) of the adhesive fracture and to account for plastic deformation of adherends. Mode I and mode II traction separation laws (TSLs) are determined from a property identification method with a Benzeggagh–Kenane mixed-mode coupling law used to model mixed-mode behavior. FEA results are shown to provide a good agreement to both the crosshead displacement and DIC data. The methods in this paper serve as a potential framework for future calibration of mixed-mode fracture envelopes for joints bonded with very tough adhesives that complicate assessment with traditional data analysis methods.more » « less
-
Understanding the fatigue behavior and accurately predicting the fatigue life of laser powder bed fusion (L-PBF) parts remain a pressing challenge due to complex failure mechanisms, time-consuming tests, and limited fatigue data. This study proposes a physics-informed data-driven framework, a multimodal transfer learning (MMTL) framework, to understand process-defect-fatigue relationships in L-PBF by integrating various modalities of fatigue performance, including process parameters, XCT-inspected defects, and fatigue test conditions. It aims to leverage a pre-trained model with abundant process and defect data in the source task to predict fatigue life nondestructive with limited fatigue test data in the target task. MMTL employs a hierarchical graph convolutional network (HGCN) to classify defects in the source task by representing process parameters and defect features in graphs, thereby enhancing its interpretability. The feature embedding learned from HGCN is then transferred to fatigue life modeling in neural network layers, enabling fatigue life prediction for L-PBF parts with limited data. MMTL validation through a numerical simulation and real-case study demonstrates its effectiveness, achieving an F1-score of 0.9593 in defect classification and a mean absolute percentage log error of 0.0425 in fatigue life prediction. MMTL can be extended to other applications with multiple modalities and limited data.more » « less
