Scholarlink

<p>This research investigates the oxygen transfer efficiency or aeration efficacy (AE) of solid jet aerators featuring circular (round) apertures, across two distinct aeration units, incorporating a diverse range of parameters: varying jet counts (from 1 to 8), different aperture areas, liquid jet lengths (ranging from 170 to 470&#xa0;mm), and flow rates (ranging from 1.05 to 3.04&#xa0;l/s). The objective is to improve the precision of AE predictions by integrating experimental data with advanced computational models, such as kernel-based Gaussian Process Regression (G.P.R.), Random Forest (R.F.), Non-Linear Regression (N.L.R.), and Support Vector Machine Regression (S.V.M.). The performance of these models was evaluated using various metrics, including Nash–Sutcliffe Efficiency (N.S.E.), Coefficient of Correlation (C.C.), Scattering Index (S.I.), Root Mean Square Error (R.M.S.E.), and Mean Absolute Error (M.A.E.). The analysis reveals that the G.P.R. model with a Radial Basis Function (R.B.F.) kernel outperforms other techniques, achieving C.C. values of 1.0000 and 0.9998, M.A.E. values of 0.034 and 0.0898, R.M.S.E. values of 0.0549 and 0.1366, N.S.E. values of 0.9999 and 0.9995 and S.I. values of 0.0092 and 0.023 for calibration and validation sets of data, respectively. Investigation of sensitivity highlights the significant impact of flow rates on AE, as evidenced by a C.C. of 0.5048, M.A.E. of 4.1747, and R.M.S.E. of 5.7473. The number of apertures plays vital role in influencing AE. This study validates the use of G.P.R. with an R.B.F. kernel as a powerful tool to refine the design of aeration systems in both effluent treatment and commercial applications.</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

Approximation of Oxygen Transfer Efficiency of Solid Jet Aerator Having Circular Opening with Kernel Function-Based Models and Random Forest Models

  • 期刊论文

  • 发表日期：

    2025年4月15日

查看全文
  • Bishnu Kant Shukla,
  • Arun Goel,
  • Pushpendra Kumar Sharma,
  • Parveen Sihag

摘要

This research investigates the oxygen transfer efficiency or aeration efficacy (AE) of solid jet aerators featuring circular (round) apertures, across two distinct aeration units, incorporating a diverse range of parameters: varying jet counts (from 1 to 8), different aperture areas, liquid jet lengths (ranging from 170 to 470 mm), and flow rates (ranging from 1.05 to 3.04 l/s). The objective is to improve the precision of AE predictions by integrating experimental data with advanced computational models, such as kernel-based Gaussian Process Regression (G.P.R.), Random Forest (R.F.), Non-Linear Regression (N.L.R.), and Support Vector Machine Regression (S.V.M.). The performance of these models was evaluated using various metrics, including Nash–Sutcliffe Efficiency (N.S.E.), Coefficient of Correlation (C.C.), Scattering Index (S.I.), Root Mean Square Error (R.M.S.E.), and Mean Absolute Error (M.A.E.). The analysis reveals that the G.P.R. model with a Radial Basis Function (R.B.F.) kernel outperforms other techniques, achieving C.C. values of 1.0000 and 0.9998, M.A.E. values of 0.034 and 0.0898, R.M.S.E. values of 0.0549 and 0.1366, N.S.E. values of 0.9999 and 0.9995 and S.I. values of 0.0092 and 0.023 for calibration and validation sets of data, respectively. Investigation of sensitivity highlights the significant impact of flow rates on AE, as evidenced by a C.C. of 0.5048, M.A.E. of 4.1747, and R.M.S.E. of 5.7473. The number of apertures plays vital role in influencing AE. This study validates the use of G.P.R. with an R.B.F. kernel as a powerful tool to refine the design of aeration systems in both effluent treatment and commercial applications.