Scholarlink

<p>This work presents the development of predictive models to estimate the sound velocity in Fatty- acid-ethyl-esters (FAEEs) under diverse conditions, employing Gradient Boosting Machine (GBM) integrated with advanced optimization strategies, including Gaussian Process Optimization (GPO), Evolutionary Strategies (ES), Batch Bayesian Optimization (BBO), and Bayesian Probability Improvement (BPI). Experimental datasets from prior studies were used for model training and validation. Among the tested approaches, the Batch Bayesian optimization model exhibited the highest predictive accuracy, achieving a test <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\:{R}^{2}=0.9988\:\)</EquationSource> </InlineEquation>and an AARE of 0.581%. The robustness of the dataset, comprising 371 experimental measurements, was verified through Monte Carlo validation. Sensitivity analysis further identified pressure as the most influential parameter governing the sound velocity in FAEEs. These results underscore the effectiveness of advanced machine learning and optimization techniques in improving the predictive understanding and pharmaceutical applications of FAEEs.</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

New insights into fatty acid ethyl esters speed of sound

  • 期刊论文

  • 发表日期：

    2026年7月1日

查看全文
  • Ahmad Adel Abu-Shareha,
  • Haider Ali Jasim Alshamary,
  • Manoj I. Patel,
  • T. Narmadha,
  • Maha Mohammed Tawfiq,
  • Vikas Wasson,
  • Murodjon Akhmadaliyev,
  • Khadjaev Yusufbek,
  • Mehrdad Mottaghi

摘要

This work presents the development of predictive models to estimate the sound velocity in Fatty- acid-ethyl-esters (FAEEs) under diverse conditions, employing Gradient Boosting Machine (GBM) integrated with advanced optimization strategies, including Gaussian Process Optimization (GPO), Evolutionary Strategies (ES), Batch Bayesian Optimization (BBO), and Bayesian Probability Improvement (BPI). Experimental datasets from prior studies were used for model training and validation. Among the tested approaches, the Batch Bayesian optimization model exhibited the highest predictive accuracy, achieving a test \(\:{R}^{2}=0.9988\:\) and an AARE of 0.581%. The robustness of the dataset, comprising 371 experimental measurements, was verified through Monte Carlo validation. Sensitivity analysis further identified pressure as the most influential parameter governing the sound velocity in FAEEs. These results underscore the effectiveness of advanced machine learning and optimization techniques in improving the predictive understanding and pharmaceutical applications of FAEEs.