电子指纹助AI以少量训练数据预测复杂合金特性

继续使用 Google、Facebook 或 Apple,即表示你同意使用条款。
电子指纹帮助 AI 在训练数据极少的情况下预测复杂合金的特性
由 Lisa Lock 编辑,Robert Egan 审阅
本文已按照 Science X 的编辑流程与政策进行审阅。编辑在确保内容可信度的同时,特别标注了以下属性:
同行评审出版物
添加到首选来源
VASt 框架概览,示例取自 HEA 设计空间。图片来源:npj Computational Materials(2026)。DOI: 10.1038/s41524-026-02187-7
发明往往取决于手头有什么材料。爱迪生的灯泡直到他能造出耐用的灯丝才真正可用;LED 灯也曾沉寂多年,直到研究人员发现了半导体的正确组合。但如果工程师无需等待合适的材料就能发明新技术,会怎样?
桑迪亚国家实验室的一个小型研究项目,或许有朝一日能帮助实现这一目标。在机器学习的辅助下,桑迪亚的计算材料科学家 Matthew Barry 及其团队正在初步探索,这最终可能成为一种识别先进金属合金的通用指南。
这样的资源可以帮助工程师创造出他们能想象的几乎任何东西;他们只需查阅如何设计具有所需性能的材料。相比之下,如今工程师通常会根据可获得的材料来限制自己的设计。
尽管迄今为止评估的合金相对较少,但早期结果令人振奋。桑迪亚团队已经证明,其模型能够快速、准确地预测复杂金属组合的性能。今年 6 月,该团队在《npj Computational Materials》杂志上报告了研究发现,指出了其模型所揭示的有用且实用的指导原则和权衡取舍。
这项研究对桑迪亚的核武器工作具有特殊重要性。理论上,设计师可以为未来的库存需求开发特种材料。
研究聚焦于开发先进合金
Barry 在 Georgia Tech 攻读研究生时为其模型开发了框架,并于 2023 年在那里获得博士学位。
来到桑迪亚后,他将自己的概念应用于对一类称为高熵合金的材料进行建模,这类合金结合了复杂且有时不寻常的金属混合物。这些合金可能非常了不起。有些合金能在高温下保持强度,而不像大多数金属那样变弱。
“高熵合金最初之所以受到关注,是因为它们的高温力学性能超过了传统的高性能高温合金,”Barry 说,“我们需要这些先进材料,才能让某些应用变得可行。”
但评估它们既困难又昂贵。计算机模拟比制作和测试实体样品更快,但仍可能太慢。
帮助研究人员发现新材料的机器学习模型已经存在,但它们在处理高熵合金时往往会损失准确度。为这些先进材料生成足够的训练数据一直是一大挑战。
但 Sandia 的模型在小数据集上表现出色。它仅用数十个样本——而非数百或数千个——进行训练,就能准确预测材料性能。这种高效性让探索新的合金组合更实用、更具成本效益,也让研究团队颇为兴奋,认为它可能很容易扩展。
“理论上它应该能够泛化到整个元素周期表;我们只是还没有尝试过,”Barry 说。
由一项巧妙的训练决策带来的效率提升
该模型最初的成功可能会引出一个问题:为什么并非所有模型都这么高效?它能如此有效的原因之一,是团队在如何教它化学知识上做出了一项巧妙的决定。
这部分需要稍作解释。
原子周围有电子环绕飞舞,就像蜜蜂围着花朵打转。当原子结合形成分子或金属晶体时,它们的电子也会结合,在“花园”周围形成一种电子群。电子会比其他区域更倾向于聚集在晶体结构的某些部位,就像蜜蜂可能偏爱某些花朵。如果你把一些“钛百合”挖走,换上“铌雏菊”,整个电子群的形态和活动也会随之改变。
Sandia 的研究人员与 Georgia Tech 和 NASA 的科学家合作,训练了他们的机器学习模型——他们将其命名为 VASt,即 Voxelized Atomic Structure(体素化原子结构)的缩写——让它通过合金独特的电子指纹来识别合金。这与通常按元素定义合金的方式有本质不同。
事实证明,这一决定很明智。该模型发现了电子分布与材料性能之间的关系,从而绕过了如果按单个元素训练所需学习的那些计算。
例如,该模型发现,钒含量越高,延展性——即材料弯曲而不折断的能力——越强,而钨则能提高强度。因此,在他们 6 月的论文中,作者提出,未来合金配方的一部分思路可能是:先加入恰好足够的钨,使材料强度满足特定应用需求,然后再尽可能多地加入钒,以防止其开裂。
需要指出的是,在那项最新研究中,Barry 关注的是热机械性能,即材料如何响应热和物理应力。研究人员认为,通过用电子而非元素来训练模型,他们能够更容易地发现趋势,并能将结果外推到含有训练数据中未包含的新元素的金属。
然而,研究人员需要更好地理解是否存在某些细微差别,使这种方法适用于高熵合金,但不一定适用于所有合金。他们也在努力通过实验验证预测,并找出模型的局限性。
尽管如此,这是一个有前景的开端,有朝一日可能帮助科学家创造出他们以前从未想到的新材料混合物,使工程师能够建造他们过去只敢梦想的机器。
Matthew C. Barry 等人,使用体素化原子结构框架的数据驱动高熵合金设计,npj Computational Materials (2026)。DOI: 10.1038/s41524-026-02187-7
期刊信息:npj Computational Materials
这篇报道背后是谁?
艺术史学士,物质文化硕士。曾任博物馆编辑、急救员和移植协调员。自 2021 年起为 Science X 担任编辑。完整简介 →
数学生物学学士,创意写作硕士。游历广泛,对科学和语言有独特见解。完整简介 →
引用:电子指纹帮助 AI 以少量训练数据预测复杂合金特性(2026 年 11 月 18 日),2026 年 11 月 18 日检索自 https://techxplore.com/news/2026-11-electron-fingerprints-ai-complex-alloy.html
本文受版权保护。除出于私人学习或研究目的的合理使用外,未经书面许可,不得以任何形式复制本文任何部分。内容仅供参考之用。
新技术通过捕捉细微的原子排布模式,提升了金属合金行为的预测精度
面向新型合金的深度学习
描绘未来:AI 解读合金微观结构,提升性能预测与设计能力
利用机器学习寻找金属的最佳配比,以制成目标合金
AI 框架融合数据与文献,加速高熵合金发现
荷叶茎与生物膜利用蒸发的水产生电能,为小型电子设备供电
微观“聚合物纤维森林”有助于为电子设备散热
迄今最快的软体机器人无需电子器件即可运行
轻如羽毛的飞行机器人自带传感器和计算机
结构电池电极使能量存储于弯曲的无人机腿部之内
AI 如何识别高频能量损耗背后的磁结构
获取即时摘要文本(Gist)
VASt 利用电子分布指纹,从数十个训练样本中预测高熵合金的热机械性能。它将钒与更高的延展性联系起来,将钨与更高的强度联系起来,从而识别出成分上的权衡。仍需进行实验验证,并在更广泛的合金类别中开展测试。
本摘要由 LLM 自动生成。完整免责声明
译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。
查看英文原文(Electron fingerprints help AI predict complex alloy properties with little training data)
By continuing with Google, Facebook or Apple, you agree to the Terms of Use.
Electron fingerprints help AI predict complex alloy properties with little training data
edited by Lisa Lock, reviewed by Robert Egan
This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:
peer-reviewed publication
Add to Preferred Sources
Overview of the VASt framework with examples from HEA design space. Credit: npj Computational Materials (2026). DOI: 10.1038/s41524-026-02187-7
Inventions often depend on what's available. Edison's light bulb did not work until he could make durable filaments; LED lights languished for years until researchers discovered the right combination of semiconductors. But what if engineers could invent new technologies without waiting for the right materials?
A small research project at Sandia National Laboratories might someday help make this possible. Aided by machine learning, Sandia computational materials scientist Matthew Barry and his team are scratching the surface of what could eventually become a universal guide to identifying advanced metal alloys.
Such a resource could help engineers create virtually anything they can imagine; they could simply look up how to design the materials with the properties they need. Today, by contrast, engineers typically constrain their designs based on the materials available to them.
Although relatively few alloys have been evaluated so far, early results have been tantalizing. The Sandia team has shown its model can quickly and accurately predict the performance of elaborate combinations of metals. In June, the team reported findings in the journal npj Computational Materials that identified useful, practical guidelines and trade-offs its model had uncovered.
The research has special importance for Sandia's nuclear weapons work. Theoretically, designers could develop exotic materials for future stockpile needs.
Research focuses on developing advanced alloys
Barry developed the framework for his model as a graduate student at Georgia Tech, where he earned a doctorate in 2023.
Since coming to Sandia, he has applied his concept to modeling a class of materials called high entropy alloys, which combine complex and sometimes unusual mixes of metals. These alloys can be remarkable. Some can stay strong at high temperatures instead of weakening like most metals do.
"High entropy alloys originally gained interest because they have high-temperature mechanical properties that exceed traditional high-performance superalloys," Barry said. "We need these advanced materials to make certain applications viable."
But evaluating them can be difficult and costly. Computer simulations are faster than making and testing physical samples, but they can still be too slow.
Machine learning models that help researchers discover new materials already exist, but they tend to lose accuracy with high entropy alloys. Generating enough training data for these advanced materials has been a major challenge.
But Sandia's model shines with small data sets. It has accurately predicted material properties after being trained on just tens of samples, not hundreds or thousands. That efficiency makes it more practical and cost-effective to explore new alloy combinations and is raising some excitement among the research team that it could scale up easily.
"It should theoretically be able to generalize across the periodic table; we just haven't tried yet," Barry said.
Efficiency driven by a clever training decision
The initial success of the model might raise the question: Why aren't all models this efficient? One reason it works so well is a clever decision the team made about how to teach it chemistry.
This part takes a little explaining.
An atom is surrounded by electrons buzzing around it like bees around a flower. When atoms combine to form molecules or metal crystals, their electrons combine, too, to form a kind of swarm around a garden. Electrons gravitate toward certain parts of the crystal structure more than others, like bees might prefer certain flowers. If you were to dig up some titanium lilies and replace them with niobium daisies, the overall shape and activity of the swarm would shift in response.
Sandia researchers, in collaboration with scientists at Georgia Tech and NASA, trained their machine learning model, which they named VASt—short for Voxelized Atomic Structure—to recognize an alloy by its unique electron fingerprint. This is fundamentally different from how alloys are normally defined, by their elements.
That decision turned out to be smart. The model discovered relationships between electron distribution and material properties, sidestepping calculations it would have needed to learn if it had been trained on individual elements.
For instance, the model found that more vanadium increases ductility, a material's ability to bend without breaking, while tungsten increases strength. So in their June paper, the authors proposed that part of a future alloy recipe might be to add just as much tungsten as needed to make a material strong enough for a particular application, then pack in as much vanadium as possible to keep it from cracking.
In that most recent study, it is important to note, Barry focused on thermo-mechanical properties, or how materials respond to heat and physical stress. The researchers believe that by training the model on electrons instead of elements, they were able to find trends more easily and could extrapolate results to metals containing new elements not included in its training data.
However, the researchers need to better understand whether there are nuances that make this approach work for high entropy alloys but not necessarily for all alloys. They are also working to validate predictions experimentally and find their model's limitations.
Still, it is a promising start that could someday help scientists create new blends of materials they had never thought of before, enabling engineers to build machines they had only dreamed of.
Matthew C. Barry et al., Data-driven design of high entropy alloys using the voxelized atomic structure framework, npj Computational Materials (2026). DOI: 10.1038/s41524-026-02187-7
Journal information:npj Computational Materials
Who's behind this story?
BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021. Full profile →
Bachelor's in mathematical biology, Master's in creative writing. Well-traveled with unique perspectives on science and language. Full profile →
Citation: Electron fingerprints help AI predict complex alloy properties with little training data (2026, November 18) retrieved 18 November 2026 from https://techxplore.com/news/2026-11-electron-fingerprints-ai-complex-alloy.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.
New technique sharpens predictions of metal alloy behavior by capturing subtle atomic patterns
Deep learning for new alloys
Mapping the future: AI deciphers alloy microstructures to enhance properties prediction and design
Using machine learning to find an optimal mixture of metals to create a desired alloy
AI framework fuses data and literature to speed high-entropy alloy discovery
Lotus stems and biofilms generate electricity from evaporating water to power small electronics
Microscopic 'polymer fiber forest' helps keep electronics cool
The fastest soft robot ever built doesn't need electronics to function
Featherweight flying robot carries its own sensors and computers
Structural battery electrode enables energy storage inside curved drone legs
How AI can identify the magnetic structures behind high-frequency energy loss
Get Instant Summarized Text (Gist)
VASt uses electron-distribution fingerprints to predict high-entropy alloy thermomechanical properties from tens of training samples. It linked vanadium to greater ductility and tungsten to greater strength, identifying composition trade-offs. Experimental validation and testing across broader alloy classes remain necessary.
This summary was automatically generated using LLM. Full disclaimer
This site uses cookies to assist with navigation, analyse your use of our services, collect data for ads personalisation and provide content from third parties. By using our site, you acknowledge that you have read and understand our Privacy Policy and Terms of Use.
原文为英文,正文由自动抓取并保留原文。
原文发布于 TechXplore(Electron fingerprints help AI predict complex alloy properties with little training data),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。
