(Press-News.org)
A research team led by Chang Keke from the Ningbo Institute of Materials Technology and Engineering (NIMTE), Chinese Academy of Sciences (CAS), has developed an innovative machine learning framework (PAMCs-MP) for predicting the mechanical properties of particle-reinforced aluminum matrix composites (PAMCs). Despite limited existing datasets, the approach uses extensive pre-training on larger aluminium alloy datasets to guide multi-objective optimization tasks effectively. The model achieves high predictive accuracy, R² values of over 92% for ultimate tensile strength and over 90% for elongation, demonstrating its robustness and reliability. The platform not only accelerates the design cycle but also offers profound insights into material behaviour, facilitating the development of high-strength, ductile aluminum composites tailored to specific application needs.
Particle reinforced aluminium matrix composites (PAMCs) exhibit exceptional potential for lightweight structural components due to their high specific strength and wear resistance. Traditionally, optimizing their complex microstructures relies heavily on extensive experimental trial-and-error methods. An approach that is very time-consuming, costly, and often limited by small datasets. Current modelling techniques, including conventional machine learning (ML) algorithms, face challenges in capturing the high-dimensional nonlinear relationships among multiple design variables such as composition, processing parameters, and particle types.
While ML models like neural networks (NNs) and decision trees have shown success in materials property prediction, their effectiveness decreases with limited data availability, common in emerging material systems. This has hindered rapid, accurate, and comprehensive optimization essential for advancing next-generation PAMCs.
To overcome these limitations, the research team developed PAMCs-MP, a transfer learning framework that pre-trains on large aluminium alloy datasets, to extract relevant features transferable to the limited PAMCs data. This transfer learning strategy enables the models to understand complex interactions between alloying elements and reinforcements, even with scarce PAMCs data, thereby significantly enhancing predictive accuracy.
PAMCs-MP was pre-trained on a dataset of 1089 Al alloy samples encompassing composition and heat treatment process parameters, followed by fine-tuning using a PAMCs dataset that integrates reinforcing particle characteristics and composite processing features. Through data preprocessing, ablation experiments on transfer strategies, and hyperparameter optimization, high-quality predictions of PAMCs' mechanical properties across a broad compositional space were realized. This addresses the long-standing challenges of insufficient accuracy and high volatility in elongation prediction encountered by traditional models.
The Future: This research establishes a machine learning framework integrating transfer learning and transformer neural networks, constructing a computational workflow that correlates the composition, processing, and mechanical properties of PAMCs. Future work will focus on the compositional optimization and inverse design of processing parameters for PAMCs, aiming to shorten the R&D cycle and reduce costs associated with traditional trial-and-error methods, ultimately enabling efficient development of PAMCs with both high strength and ductility.
The Impact: This work provides a feasible technical pathway for property prediction and material design in complex material systems constrained by small datasets. By transferring knowledge from related material systems, the framework can be systematically extended to the optimization of other novel structural materials, accelerating innovations in high-performance lightweight materials for critical industries.
The research has been recently published in the online edition of Materials Futures, a prominent international journal in the field of interdisciplinary materials science research.
Reference: Qingtao Jia, Kai Xu, Changheng Li, Gaohui Kan, Yanyu Liu, Hui Ren, Shuai Zhang, Ming Lou, Keke Chang. Transfer Learning-Assisted Multi-Objective Optimization of Mechanical Properties for Particle Reinforced Aluminum Matrix Composites[J]. Materials Futures. DOI: 10.1088/2752-5724/ae2347
END
Catalytic dry reforming of methane (DRM) is a key reaction for the sustainable utilization of major greenhouse gases, CO2 and CH4. However, conventional DRM often suffers from severe catalyst deactivation due to high temperature requirements. Applying direct current (DC) to catalyst materials has emerged as a promising strategy to overcome these limitations, yet the underlying DC-enhanced catalytic mechanism remains elusive. Here, we unveil the non-thermal catalytic origin of DC-applied DRM over Pd/CeO2 through multimodal operando analyses, providing a microscopic physicochemical framework ...
Chronic breathlessness, a symptom often overlooked by healthcare systems, is associated with longer lengths of hospital stay on already overstretched healthcare resources, says new Flinders University research.
A new study, published in the Australian Health Review, highlights an urgent need for clinicians and policymakers to recognise chronic breathlessness as a major driver of hospital admissions and healthcare costs.
Historically, health systems have focused on sudden, short-term episodes of breathlessness (acute breathlessness), ...
Key points
Paleontologists working in a cave on the Caribbean island of Hispaniola have discovered the first-known instance of ancient bees nesting inside pre-existing fossil cavities.
Burrowing bees generally prefer to make their nests out in the open. There is only one other documented case of burrowing bees making their nests inside caves. In this case, the likely cause for this aberrant behavior is a lack of topsoil outside the cave and an abundance of accumulated silt within.
Many of the fossils were likely transported to the cave by giant barn owls. Evidence, ...
About 20,000 years ago, a family of owls lived in a cave. Sometimes, they would cough up owl pellets containing the bones of their prey, which landed on the cave floor. And, researchers have just discovered, ancient bees would use the bones’ empty tooth sockets as nests. A new study published in the journal Royal Society Open Science documents this discovery, which represents the first time bees have ever been known to use bones as places to lay their eggs.
The Caribbean island of Hispaniola, which contains Haiti and the Dominican Republic, is full of limestone caves. “In some areas, you’ll find a different sinkhole every 100 meters,” says ...
Are you a social savant who easily reads people’s emotions? Or are you someone who leaves an interaction with an unclear understanding of another person’s emotional state?
New UC Berkeley research suggests those differences stem from a fundamental way our brains compute facial and contextual details, potentially explaining why some people are better at reading the room than others — sometimes, much better.
Human brains use information from faces and background context, such as the location or expressions ...
New Australian research has identified simultaneous abnormalities across multiple biological systems in people with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS).
Key findings of a multimodal study published today in the journal Cell Reports Medicine include changes in markers of cellular energy metabolism, in the proportions and maturity of circulating immune cells, and in plasma proteins associated with blood vessel dysfunction in people with ME/CFS.
Led by researchers from Macquarie University, the study compared whole blood samples from 61 people meeting ...
Concurrent physical frailty and depression likely boost the risk of dementia in older people, with the interaction of these 2 factors alone contributing around 17% of the overall risk, suggest the findings of a large international study, published in the open access journal General Psychiatry.
Globally, some 57 million people are living with dementia—a figure that is expected to triple by 2050, note the researchers.
Previously published research has primarily focused on the individual associations between physical frailty or depression and dementia risk, despite the fact ...
Children living in substandard housing in England miss 15 more school days and achieve worse test scores in English and maths than their peers living in better quality housing, suggests research published online in the Journal of Epidemiology & Community Health.
Improving their living conditions—specifically reducing overcrowding and damp, and upgrading heating systems—may not only benefit their health, but also their grades,conclude the researchers.
One in 7 families in England live in homes that fail to meet the official decent homes standard, point out the researchers. Housing is a key determinant of child health, yet relatively little is ...
There is little awareness, particularly among clinicians, of the medical and psychological complexities of ‘topical steroid withdrawal’—the body’s adverse response to the prolonged use of these powerful creams to treat inflammatory skin conditions when they are either tapered or suddenly stopped—warn doctors in the journal BMJ Case Reports.
The condition, also known as ‘TSW syndrome,’ ‘steroid addiction,’ and ‘red burning skin syndrome,’ is poorly ...
Most (79%) of NHS trusts in England are treating patients in corridors or makeshift areas in emergency departments including “fit to sit” rooms, x-ray waiting areas, and in one case a café, finds an investigation published by The BMJ today.
Data obtained by The BMJ show that such practices have resulted in at least half a million patients being cared for in temporary spaces and that in some trusts one in four patients in accident and emergency (A&E) departments were cared for in corridors last year.
Corridor care refers to the practice of providing care to patients ...
