基于Aspen Plus和决策树的车间培训数字孪生系统构建  

Construction of Digital Twin System for Workshop Training Based on Aspen Plus and Decision Tree

作　　者：何仲坤 单宝明[1] HE Zhong-kun;SHAN Bao-ming(College of Automation and Electronic Engineering,Qingdao University of Science&Technology)

机构地区：[1]青岛科技大学自动化与电子工程学院

出　　处：《化工自动化及仪表》2024年第4期706-713,共8页Control and Instruments in Chemical Industry

摘　　要：针对传统化工企业车间操作培训方式教学效果难以量化、费用高昂、时间成本大等问题,研究了以数字孪生技术为基础的化工企业车间操作培训系统和自动评测模型。采用不同领域的仿真软件相互连接来模拟整个流程,通过可视化结果辅助数据分析和决策过程,验证了Aspen Plus软件在化工数字孪生领域应用的可行性。此外,还提出了一种基于决策树和Dijkstra算法的自动评测模型,利用决策树对数据进行分类,结合Dijkstra算法计算最优标准序列,实现了针对员工操作过程和结果的自动评估。Considering traditional training methods’difficulty in quantifying the teaching effect and their high cost and long time consumption for chemical workshop operation,a workshop training system and auto-assessment model for chemical enterprises based on the digital twin technology were studied.Through having the simulation software from different fields connected each other to simulate the whole process,and having visualizing results adopted to assist the data analysis and decision-making process,the feasibility of Aspen Plus software applied in chemical digital twin was verified.In addition,an auto-evaluation model based on decision tree and Dijkstra algorithm was proposed to classify the data and calculate the optimal criteria sequence to achieve auto-evaluation of employee operation process and results.

关 键 词：数字孪生 操作培训系统 自动评测模型 决策树 DIJKSTRA 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]