Scholarlink

<p>An accurate study of extreme precipitation risk events is crucial for flood control, urban planning, and engineering design. The Copula function effectively handles uncertainties and nonlinear interactions among multiple hydrological variables, capturing complex correlations in extreme precipitation events for more precise risk assessments. Selecting the parameters of the Copula function is critical, as it defines the function’s shape and the dependence structure between variables, influencing its application. Traditional parameterization methods, like maximum likelihood estimation and least squares, often require large datasets and distribution assumptions, making them cumbersome for high-dimensional data. This research presents a model using an enhanced whale optimization algorithm to estimate Copula parameters (CLCWOA-Copula), aiming to assess return period (RP) and failure probability associated with extreme precipitation risk events. Monthly precipitation data from Zhengzhou, China, from 1950 to 2024 is analyzed, using the 90th percentile as the extreme precipitation threshold. Marginal distributions are fitted using the P-III and gamma distributions, etc., which are then combined using CLCWOA-Copula to analyze coincident RP, joint RP, Kendall RP, and failure probability under composite conditions. The results demonstrate that this optimization method possesses strong global search capabilities and parallel computing abilities, yielding optimal Copula parameter values within few iterations and selecting the best-fit Copula function based on AIC, <i>R</i>², and RMSE. The Kendall RP and failure probability offer more accurate tools for extreme precipitation risk analysis; when P<sub>max</sub> reaches 540&#xa0;mm, P<sub>90</sub> reaches 1080&#xa0;mm, or R<sub>90</sub> reaches 0.83, a one-in-a-century extreme precipitation event is indicated. This study provides important insights for risk metrics applicable to extreme weather.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Extreme precipitation risk assessment with improved WOA-Optimized copula model under composite conditions

  • 期刊论文

  • 发表日期：

    2025年8月4日

查看全文
  • Zhaocai Wang,
  • Chao Ma,
  • Junhao Wu,
  • Tunhua Wu

摘要

An accurate study of extreme precipitation risk events is crucial for flood control, urban planning, and engineering design. The Copula function effectively handles uncertainties and nonlinear interactions among multiple hydrological variables, capturing complex correlations in extreme precipitation events for more precise risk assessments. Selecting the parameters of the Copula function is critical, as it defines the function’s shape and the dependence structure between variables, influencing its application. Traditional parameterization methods, like maximum likelihood estimation and least squares, often require large datasets and distribution assumptions, making them cumbersome for high-dimensional data. This research presents a model using an enhanced whale optimization algorithm to estimate Copula parameters (CLCWOA-Copula), aiming to assess return period (RP) and failure probability associated with extreme precipitation risk events. Monthly precipitation data from Zhengzhou, China, from 1950 to 2024 is analyzed, using the 90th percentile as the extreme precipitation threshold. Marginal distributions are fitted using the P-III and gamma distributions, etc., which are then combined using CLCWOA-Copula to analyze coincident RP, joint RP, Kendall RP, and failure probability under composite conditions. The results demonstrate that this optimization method possesses strong global search capabilities and parallel computing abilities, yielding optimal Copula parameter values within few iterations and selecting the best-fit Copula function based on AIC, R², and RMSE. The Kendall RP and failure probability offer more accurate tools for extreme precipitation risk analysis; when Pmax reaches 540 mm, P90 reaches 1080 mm, or R90 reaches 0.83, a one-in-a-century extreme precipitation event is indicated. This study provides important insights for risk metrics applicable to extreme weather.