کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
5766362 1627734 2017 19 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Adaptive subdomain modeling: A multi-analysis technique for ocean circulation models
ترجمه فارسی عنوان
مدل سازی زیر دامنه سازگار: یک تکنیک چند تجزیه ای برای مدل های گردش اقیانوس
موضوعات مرتبط
مهندسی و علوم پایه علوم زمین و سیارات علم هواشناسی
چکیده انگلیسی


- A technique is presented that reduces the time needed to assess local domain changes.
- These may correspond to design alternatives and failure scenarios to be simulated.
- Differences of interest are incorporated in subdomains with moving boundaries.
- The spatial extent of each subdomain is adaptively adjusted during runtime.
- The technique works by concurrently analyzing the differences in model responses.

Many coastal and ocean processes of interest operate over large temporal and geographical scales and require a substantial amount of computational resources, particularly when engineering design and failure scenarios are also considered. This study presents an adaptive multi-analysis technique that improves the efficiency of these computations when multiple alternatives are being simulated. The technique, called adaptive subdomain modeling, concurrently analyzes any number of child domains, with each instance corresponding to a unique design or failure scenario, in addition to a full-scale parent domain providing the boundary conditions for its children. To contain the altered hydrodynamics originating from the modifications, the spatial extent of each child domain is adaptively adjusted during runtime depending on the response of the model. The technique is incorporated in ADCIRC++, a re-implementation of the popular ADCIRC ocean circulation model with an updated software architecture designed to facilitate this adaptive behavior and to utilize concurrent executions of multiple domains. The results of our case studies confirm that the method substantially reduces computational effort while maintaining accuracy.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Ocean Modelling - Volume 115, July 2017, Pages 86-104
نویسندگان
, ,