GA-Based Design Method for Driven Piles in Alberta

艾伯塔省基于遗传算法的打入桩设计方法

• 批准号: 539280-2019
• 负责人: Liu, Jinyuan
• 金额: $ 1.82万
• 依托单位: Ryerson University
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2019
• 资助国家: 加拿大
• 起止时间: 2019-01-01 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=680734
• 关键词: GA; Based; Design; Method; Driven

This proposed research is to develop a new design method for driven piles in Alberta soils using the Genetic Algorithm (GA), a machine learning technique. Pile foundations are frequently used to support new structures or upgrade existing infrastructure throughout Canada. There are many inconsistencies in existing design methods due to many uncertainties and variables involved in the design and construction. Based on a database of more than 100 PDA load tests collected in Alberta, this study will correlate the pile resistances with soil parameters using GA and develop a reliability-based GA design method for driven piles in Alberta soils. This project helps support cost-effective infrastructure developments in Alberta. This project will crucially involve the training of highly qualified personnel to solve real engineering problems.

本文提出的研究是利用遗传算法（GA），一种机器学习技术，开发一种新的阿尔伯塔省土壤中打入桩的设计方法。桩基在加拿大各地经常用于支持新结构或升级现有基础设施。由于设计和施工过程中存在许多不确定因素和变量，现有的设计方法存在许多不一致性。基于在Alberta地区收集的100多个PDA荷载试验数据，本研究将利用遗传算法将桩阻力与土体参数关联起来，并建立基于遗传算法的Alberta土中打入桩可靠性设计方法。该项目有助于支持阿尔伯塔省具有成本效益的基础设施发展。这个项目的关键在于培养高素质的人才来解决实际的工程问题。