Scholarlink

<p>Artificial intelligence offers a powerful way to understand and predict the flow and thermal behavior of non-Newtonian fluids with greater accuracy and efficiency. This study presents a comprehensive AI-based numerical analysis of an Oldroyd-B nanofluid within a porous medium by incorporating the effects of entropy generation. The nanofluids consist of silver (Ag) and nickel zinc ferrite (NiZnFe<sub>2</sub>O<sub>4</sub>) nanoparticles dispersed in engine oil, which is the base fluid. The thermal convective boundary conditions as well as velocity slip significantly affect the fluid’s transport phenomena. The differential transform method (DTM) is employed to obtain approximate solutions, whereas a Bayesian regression neural network (BRNN) framework is utilized to perform the artificial intelligence-based predictive and comparative analysis. The artificial intelligence model is trained using the computed numerical data. This model describes the insights into the different governing parameters that influence the flow and thermal characteristics. An increase in nanoparticle volume fraction ϕ enhances the Nusselt number by nearly 18–30%, with the Ag/engine oil nanofluid consistently outperforming the NiZnFe<sub>2</sub>O<sub>4</sub>/engine oil nanofluid. The BRNN predictive model achieves regression values exceeding 0.99, indicating high accuracy. The outcomes of this research hold significant relevance for advancements in nanotechnology, thermal engineering, and materials science where efficient heat and fluid transport are of critical importance.</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

Artificial intelligence-based Bayesian regression predictions and entropy generation analysis of Oldroyd-B fluid flow through porous structures

  • 期刊论文

  • 发表日期：

    2026年4月20日

查看全文
  • Sohail Ahmad,
  • Muhammad Amjad,
  • Ghulam Murtaza,
  • Gulnaz Atta,
  • Maimoona Karim,
  • Zehba Raizah,
  • Hammad Alotaibi

摘要

Artificial intelligence offers a powerful way to understand and predict the flow and thermal behavior of non-Newtonian fluids with greater accuracy and efficiency. This study presents a comprehensive AI-based numerical analysis of an Oldroyd-B nanofluid within a porous medium by incorporating the effects of entropy generation. The nanofluids consist of silver (Ag) and nickel zinc ferrite (NiZnFe2O4) nanoparticles dispersed in engine oil, which is the base fluid. The thermal convective boundary conditions as well as velocity slip significantly affect the fluid’s transport phenomena. The differential transform method (DTM) is employed to obtain approximate solutions, whereas a Bayesian regression neural network (BRNN) framework is utilized to perform the artificial intelligence-based predictive and comparative analysis. The artificial intelligence model is trained using the computed numerical data. This model describes the insights into the different governing parameters that influence the flow and thermal characteristics. An increase in nanoparticle volume fraction ϕ enhances the Nusselt number by nearly 18–30%, with the Ag/engine oil nanofluid consistently outperforming the NiZnFe2O4/engine oil nanofluid. The BRNN predictive model achieves regression values exceeding 0.99, indicating high accuracy. The outcomes of this research hold significant relevance for advancements in nanotechnology, thermal engineering, and materials science where efficient heat and fluid transport are of critical importance.