CAREER: Soil Liquefaction Evaluations at Multiple Scales: Reshaping Research, Training, and Education Through Physics-Guided Data Science

职业：多尺度土壤液化评估：通过物理引导数据科学重塑研究、培训和教育

• 批准号: 2047838
• 负责人: Katerina Ziotopoulou
• 金额: $ 56.78万
• 依托单位: University of California-Davis
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-07-01 至 2026-06-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2047838&HistoricalAwards=false
• 关键词: CAREER; Soil; Liquefaction; Evaluations; Multiple

This Faculty Early Career Development (CAREER) award will investigate and formalize new approaches for learning from geotechnical data, and inform the next generation of liquefaction evaluation tools. Soil liquefaction remains a leading cause of earthquake damage worldwide. The rarity of earthquakes poses an enduring challenge, geosystems are complex and hard to sample, and the uniqueness of each requires predictive inferences beyond the existing database. Owing to cutting-edge instrumentation and experimentation, new reconnaissance approaches, and generous resources for data storage, the increasing availability of geotechnical data, combined with open-source algorithms, creates an opportunity to exploit the potential of Data Science for reshaping our engineering toolbox. Aligned with NSF’s Big Ideas on Harnessing the Data Revolution and Convergence Research, this CAREER award will open new avenues for integrating physics-based and data-driven methods to accelerate discoveries in geotechnical engineering and develop robust validated geotechnical tools that will facilitate integration with multi-hazard analysis capabilities and broadly contribute to reducing earthquake risk and increasing societal resilience. The standards of practice will also be elevated by formulating efficient test programs that meet economic constraints. Through an organic education plan, this CAREER grant will prepare the geotechnical engineers for the workplace of the future by fostering four core competencies: (i) interdisciplinary thinking to achieve creative solutions in research and practice, (ii) Data Science literacy to harness the power of engineering data, communicate information visually, and deliver reproducible, and transparent products, (iii) diversity to develop geoengineering solutions on a wealth of perspectives and experiences, and (iv) communication skills to engage in teamwork and effective dissemination. All data will be archived and made publicly available in the Natural Hazards Engineering Research Infrastructure (NHERI) Data Depot.The research aims to investigate, prove, and formalize the application of Physics-Guided Data Science in liquefaction evaluations by working at three scales with a wealth of numerical and experimental data. The objectives are to probe known, and discover new, high-dimensional interactions of soil properties, loading, and responses for liquefiable soils at the element and centrifuge model scale, and explore the transfer of knowledge to a field case history. The scope includes: (i) development of transferable mechanics-based engineering models, (ii) quantitative validation framework for numerical models, (iii) methodology for reduction of suspected biases in experimental and numerical data, (iv) smarter experimental design targeting key predictors for responses of interest, and (v) smarter site investigation campaigns to identify risk-driving vulnerabilities. The T-shaped education plan will develop pedagogical approaches that include: (i) case-based student-centered hybrid teaching modules, (ii) interactive user experiences introducing students to basic concepts of earthquake engineering and data-enabled decision making, (iii) training and support of Earth Science teachers from underserved districts, (iv) outreach to pre-middle school girls and ladder mentoring of female students to address diversity and inclusion, and (v) establishment of a new series of department- and college-wide activities investigating interdisciplinary thinking. This CAREER grant will also provide summer support to three graduate students working in Hazards Engineering to formalize interdisciplinary conversations and accordingly pursue innovative research avenues.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.

这个教师早期职业发展（CAREER）奖将调查和正式化从岩土数据学习的新方法，并告知下一代液化评估工具。土壤液化仍然是全球地震破坏的主要原因。地震的罕见性构成了一个持久的挑战，地理系统是复杂的，很难采样，每一个的独特性需要超越现有数据库的预测推理。由于尖端的仪器和实验，新的勘测方法以及用于数据存储的慷慨资源，岩土数据的可用性不断增加，再加上开源算法，为利用数据科学的潜力重塑我们的工程工具箱创造了机会。与NSF关于利用数据革命和融合研究的大想法相一致，该职业奖将为整合基于物理和数据驱动的方法开辟新的途径，以加速岩土工程的发现，并开发强大的验证岩土工具，以促进与多灾害分析能力的集成，并广泛有助于降低地震风险和提高社会复原力。实践标准也将通过制定符合经济限制的有效测试计划来提高。通过一个有机的教育计划，这个职业生涯补助金将通过培养四个核心能力为未来的工作场所做好准备岩土工程师：（i）跨学科思维，以实现研究和实践中的创造性解决方案，（ii）数据科学素养，以利用工程数据的力量，直观地传达信息，并提供可复制和透明的产品，（iii）以丰富的观点和经验开发地球工程解决方案的多样性，以及（iv）参与团队合作和有效传播的沟通技能。所有数据将在自然灾害工程研究基础设施（NHERI）数据库中存档并公开提供。该研究旨在通过三个尺度的大量数值和实验数据，调查，证明和正式化物理指导的数据科学在液化评估中的应用。目的是探索已知的，并发现新的，高维的土壤性质，负载和响应的相互作用，可液化土壤在元素和离心模型的规模，并探讨知识转移到现场的案例历史。范围包括：（i）开发可转移的基于力学的工程模型，（ii）数值模型的定量验证框架，（iii）减少实验和数值数据中可疑偏差的方法，（iv）针对感兴趣的反应的关键预测因素的更智能的实验设计，以及（v）更智能的现场调查活动，以确定风险驱动的脆弱性。T形教育计划将制定教学方法，包括：（i）以学生为中心的案例混合教学模块，（ii）互动用户体验，向学生介绍地震工程和数据决策的基本概念，（iii）培训和支持服务不足地区的地球科学教师，（iv）向中学前女生开展外联活动，并对女学生进行阶梯式辅导，以解决多样性和包容性问题，以及（v）在整个系和学院范围内开展一系列新的活动，调查跨学科思维。这项CAREER资助还将为三名从事危险工程的研究生提供暑期支持，以正式开展跨学科对话，并相应地追求创新的研究途径。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Machine Learning Applications in Geotechnical Earthquake Engineering: Progress, Gaps, and Opportunities

• DOI: 10.1061/9780784484692.050
• 发表时间: 2023-03
• 期刊: Geo-Congress 2023
• 作者: Katherine Cheng;K. Ziotopoulou