面向化工过程仿真的层次化任务图并行计算框架

, , , , ,

工程(英文) ›› 2025, Vol. 51 ›› Issue (8) : 229 -239.

PDF
工程(英文) ›› 2025, Vol. 51 ›› Issue (8) : 229 -239. DOI: 10.1016/j.eng.2024.06.019
研究论文

面向化工过程仿真的层次化任务图并行计算框架

作者信息 +

A Hierarchical Task Graph Parallel Computing Framework for Chemical Process Simulation

Author information +
文章历史 +
PDF

Abstract

Sequential-modular-based process flowsheeting software remains an indispensable tool for process design, control, and optimization. Yet, as the process industry advances in intelligent operation and maintenance, conventional sequential-modular-based process-simulation techniques present challenges regarding computationally intensive calculations and significant central processing unit (CPU) time requirements, particularly in large-scale design and optimization tasks. To address these challenges, this paper proposes a novel process-simulation parallel computing framework (PSPCF). This framework achieves layered parallelism in recycling processes at the unit operation level. Notably, PSPCF introduces a groundbreaking concept of formulating simulation problems as task graphs and utilizes Taskflow, an advanced task graph computing system, for hierarchical parallel scheduling and the execution of unit operation tasks. PSPCF also integrates an advanced work-stealing scheme to automatically balance thread resources with the demanding workload of unit operation tasks. For evaluation, both a simpler parallel column process and a more complex cracked gas separation process were simulated on a flowsheeting platform using PSPCF. The framework demonstrates significant time savings, achieving over 60% reduction in processing time for the simpler process and a 35%–40% speed-up for the more complex separation process.

关键词

Key words

Parallel computing / Process simulation / Task graph parallelism / Sequential modular approach

引用本文

引用格式 ▾
, , , , , 面向化工过程仿真的层次化任务图并行计算框架[J]. 工程(英文), 2025, 51(8): 229-239 DOI:10.1016/j.eng.2024.06.019

登录浏览全文

4963

注册一个新账户 忘记密码

参考文献

AI Summary AI Mindmap
PDF

Supplementary files

Supplementary data

456

访问

0

被引

详细

导航
相关文章

AI思维导图

/