CAREER: A Non-local Mathematical and Computational Paradigm for Failure in Unsaturated Soils: Integrated Research and Education through High Performance Computing

该学院早期职业发展计划（Career）赠款将解决基础知识差距，这些知识差距阻止了对驱动非饱和土壤失效（例如剪切带/裂缝）的多物理场条件的准确表征。非饱和土力学在地质灾害评价以及能源、环境和海岸岩土工程中起着至关重要的作用。这些领域的进展受到非饱和土破坏机制的阻碍，这些机制特征不佳，因此难以预测和缓解。这种破坏涉及多物理场耦合过程和多空间、多时间尺度的任意剪切带/裂缝。随着高性能计算（HPC）的最新进展，计算建模对于提高我们表征、预测和减轻此类故障的能力变得越来越重要。这个强大的工具必须与新的和强大的数值方法相结合，旨在利用和优化这种能力。本CAREER项目将开发一种新的非局部数学和计算范式，用于通过HPC模拟非等温非饱和土壤的破坏现象。假设材料的非均质性和环境荷载是非饱和土中破坏（剪切带/裂缝）的关键触发因素，并且可以通过将HPC与新的基于物理的数值工具相结合来实现这种破坏的预测。非饱和土力学、界面物理、孔隙力学、热力学、非局部矢量微积分和高性能计算的严格整合可能会彻底改变我们对多尺度、多物理场问题的建模技术。通过HPC进行的综合研究和教育活动将培养高中和代表性不足的学生对STEM教育和职业的兴趣，并吸引研究生参与全球合作研究，并向不同的受众有效传播科学知识。包括美国国家科学基金会中心、美国国家实验室、领先的咨询公司和全球顶级机构在内的多元化合作网络，将增加该项目在国内和全球的影响，并使学生接触到一个充满活力的团队。本CAREER基金的研究目标是更好地表征和预测环境荷载下非饱和土壤的破坏。该项目将(i)使用非局部矢量微积分和物理力学基本原理制定、实施和验证一种新的多物理场孔隙力学（PPM）范式，以及（ii）通过HPC进行广泛的计算实验。该项目产生的新知识包括对驱动非饱和土壤失效的多物理场条件的基本机制理解，这对于建设可持续和有弹性的民用基础设施至关重要。该项目的一个重要成果预计是一种新的多物理场PPM范式，由于其在多个空间尺度上的物理和数学一致性，该范式可能会改变非饱和土壤失效的数学和计算模型。预期的原始贡献是：(i)为更好地模拟土壤多物理场提供一种新的物理、数学和计算一致的范式；（ii）下一代非饱和土非局部本构模型；（iii）用于预测非饱和土破坏的经过验证的开源HPC工具；（iv）对非饱和土中驱动失效的多物理场条件的新机制理解。该教育计划将综合数学、物理、力学、高性能计算（HPC）的基本原理，培养具有多样化知识基础的下一代工程师和科学家。该项目将实施一系列举措，包括两个课程模块，介绍非饱和土破坏分析的关键概念和高性能计算在基础科学研究中的重要作用，一个云计算应用程序，一个专门的wiki页面，以及NSF SimCenter和ASCE网络研讨会。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This Faculty Early Career Development Program (CAREER) grant will address fundamental knowledge gaps preventing the accurate characterization of multiphysics conditions driving failure (e.g., shear bands/cracks) of unsaturated soils. Unsaturated soil mechanics plays a vital role in geohazard assessment, as well as energy, environmental, and coastal geotechnics. Advances in these fields are hindered by unsaturated soil failure mechanisms that are poorly characterized and thus difficult to predict and mitigate. Such failures involve coupled multiphysics processes and arbitrary shear bands/cracks at multiple space and time scales. With recent advances in high-performance computing (HPC), computational modeling is becoming increasingly crucial for advancing our ability to characterize, predict, and mitigate such failures. This powerful tool must be coupled with new and robust numerical methods that are designed to harness and optimize this capability. This CAREER project will develop a novel non-local mathematical and computational paradigm for modeling failure phenomena in non-isothermal unsaturated soils through HPC. The hypothesis is that material heterogeneities and environmental loads are the critical triggers for failures (shear bands/cracks) in unsaturated soils, and that the prediction of such failures can be achieved via coupling HPC with new physics-based numerical tools. The rigorous integration of unsaturated soil mechanics, interface physics, poromechanics, thermodynamics, non-local vector calculus, and HPC can potentially revolutionize our modeling techniques for multiscale, multiphysics problems. Integrated research and educational activities through HPC will foster the interest of high-school and underrepresented students in STEM educations and careers and engage graduate students in globally collaborative research and effective dissemination of scientific knowledge to a diverse audience. The cultivated diverse collaboration network, including NSF Centers, U.S. national laboratories, leading consulting firms, and top global institutions, will increase the national and global impacts of this project and has the added benefit of exposing students to a dynamic team.The research goal of this CAREER grant is to better characterize and predict failures in unsaturated soils under environmental loads. This project will (i) formulate, implement, and validate a novel multiphysics peri-poromechanics (PPM) paradigm using non-local vector calculus and basic principles of physics and mechanics, and (ii) conduct extensive computational experiments through HPC. The new knowledge generated by this project includes a fundamental mechanistic understanding of multiphysics conditions driving failures of unsaturated soils that is crucial for building sustainable and resilient civil infrastructure. A significant outcome of this project is expected to be a novel multiphysics PPM paradigm that can potentially transform mathematical and computational modeling of failures in unsaturated soils due to its physical and mathematical consistency across multiple spatial scales. Original contributions expected are: (i) A novel physically, mathematically and computationally consistent paradigm for better modeling soil multiphysics; (ii) Next-generation non-local constitutive models for unsaturated soils; (iii) A validated open-source HPC tool for predicting unsaturated soil failures; and (iv) A new mechanistic understanding of multiphysics conditions driving failures in unsaturated soils. The educational plan will challenge and prepare next-generation engineers and scientists with a diverse knowledge base by integrating fundamental principles of mathematics, physics, mechanics, and HPC. This project will implement a series of initiatives, including two course modules on key concepts in unsaturated soil failure analysis and the vital role of HPC in basic scientific research, a cloud computing app, a dedicated wiki page, and NSF SimCenter and ASCE webinars.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.