Scholarlink

The study aims to enhance the processing of hydrocarbon and chemical feedstocks by adjusting technological parameters or feedstock composition. The research involved modelling a gas fractionation unit of an operational plant using the universal simulation software Aspen Hysys V12. Based on the modelling results, a machine learning model was developed in the cloud-based programming environment Google Colaboratory using Python and built-in libraries Pandas and Sklearn for data analysis and machine learning. The model predicts changes in the yield of the target product depending on variations in the absorption unit’s operating parameters, specifically the absorbent flow rate and feedstock composition. The obtained results improve the efficiency of the absorption process in gas fractionation units. The prediction process is based on the linear regression method, achieving a forecasting accuracy of 99.4%. The developed model serves as a universal monitoring tool to maximise target product yield in gas fractionation units, making it valuable both technically and economically.

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

Optimization of Gas Fractionation Processes: Integration of Aspen Hysys Simulation and Machine Learning for Enhanced Ethylene Recovery and Process Efficiency

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • V. V. Bronskaya,
  • A. V. Shipin,
  • M. I. Kondrateva,
  • A. A. Firsin,
  • A. A. Goryachev

摘要

The study aims to enhance the processing of hydrocarbon and chemical feedstocks by adjusting technological parameters or feedstock composition. The research involved modelling a gas fractionation unit of an operational plant using the universal simulation software Aspen Hysys V12. Based on the modelling results, a machine learning model was developed in the cloud-based programming environment Google Colaboratory using Python and built-in libraries Pandas and Sklearn for data analysis and machine learning. The model predicts changes in the yield of the target product depending on variations in the absorption unit’s operating parameters, specifically the absorbent flow rate and feedstock composition. The obtained results improve the efficiency of the absorption process in gas fractionation units. The prediction process is based on the linear regression method, achieving a forecasting accuracy of 99.4%. The developed model serves as a universal monitoring tool to maximise target product yield in gas fractionation units, making it valuable both technically and economically.