Complexity to clarity: detecting, identifying and analysing complex materials systems at the molecular level with machine learning

从复杂到清晰：通过机器学习在分子水平上检测、识别和分析复杂材料系统

• 批准号: NI210100127
• 负责人: Prof Paul Pigram
• 金额: $ 44万
• 依托单位: La Trobe University
• 依托单位国家: 澳大利亚
• 项目类别: National Intelligence and Security Discovery Research Grants
• 财政年份: 2021
• 资助国家: 澳大利亚
• 起止时间: 2021-12-20 至 2024-12-19
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/NI210100127
• 关键词: Complexity; clarity; detecting; identifying; analysing

This project aims to develop machine learning methodologies to characterise, classify and predict the properties of complex materials systems including nanostructured functional surfaces, multilayers and coatings. This project expects to generate a framework for deep level and precise forensic investigation of materials type, source, origin and manufacture, employing self organising maps and innovative hyperspectral data visualisation. Expected outcomes of this project include enhanced capacity to investigate unknown materials, deconstruct complex failures and degradation processes and predict structure - function relationships. This will provide significant benefits in addressing threats from emerging materials technologies and processes.

该项目旨在开发机器学习方法来表征、分类和预测复杂材料系统的性能，包括纳米结构功能表面、多层膜和涂层。该项目预计将产生一个框架，利用自组织地图和创新的高光谱数据可视化，对材料类型、来源、来源和制造进行深层次和精确的法医调查。该项目的预期成果包括增强调查未知材料、解构复杂的故障和降解过程以及预测结构-功能关系的能力。这将在应对来自新兴材料、技术和工艺的威胁方面带来重大好处。