极值理论
计算机科学
统计模型
概率逻辑
可靠性(半导体)
数据科学
多元统计
运筹学
风险分析(工程)
管理科学
计量经济学
统计
工程类
机器学习
人工智能
数学
医学
量子力学
物理
功率(物理)
作者
Erik Vanem,Tingyao Zhu,Alexander V. Babanin
标识
DOI:10.1016/j.marstruc.2022.103297
摘要
Probabilistic modelling and statistical analysis of environmental conditions is important for the design and assessment of ships and other marine structures. It will give a necessary input to structural reliability assessments and risk analysis and provides a means to identify design conditions that structures are expected to withstand in their lifetime. In this paper, recent developments in statistical modelling of relevant metocean variables describing the environment at sea will be reviewed and presented. This includes a review of statistical modelling applied to such data, for example wave-parameters, but also some theoretical and methodological developments from other fields of applications will be reviewed. The paper is divided into different sub-sections addressing various aspects of statistical modelling of the ocean environment, such as long-term and short-term statistics, extreme value analysis, non-stationary analysis and covariate effects, multivariate analysis and joint distributions, spatial statistics and machine learning applications. This distinction into sub-topics may be somewhat arbitrary, and some papers address several of these topics, e.g. non-stationary, multivariate extreme value statistics for spatial data, but it is believed to be useful to still keep separate sections for the main aspects. It is believed that this overview of recent developments in statistical modelling of the ocean environment will be useful for anyone involved in design or risk assessment of marine structures, and that it may contribute to push the state of the art and industry practice.
科研通智能强力驱动
Strongly Powered by AbleSci AI