Scholarlink

<p>The present study is backed by prior research from the authors (Patel et al. 2020), which further continued aiming to optimize and model the adsorption process for greywater treatment using adsorbents derived from pine needles, sugarcane bagasse, and sawdust. Initial analysis of greywater characteristics revealed concentrations of chemical oxygen demand (COD) and biochemical oxygen demand (BOD) as 554 and 120&#xa0;mg/L, respectively. In batch adsorption experiments, sawdust activated carbon (SDAC) exhibited the highest efficiency, in removing 97.83% of COD and 95.83% of BOD, followed by sugarcane bagasse activated carbon (SCBAC) (91.85% of COD and 90% of BOD) and pine needle activated carbon (PNAC) (95.30% of COD and 93.33% of BOD), under operating condition (pH 7, dosage: 8&#xa0;g/L, contact time: 240&#xa0;min). An approach combining response surface methodology (RSM) with artificial neural network (ANN) was employed for the optimization of batch adsorption parameters and to address their individual limitations. The feasibility of the developed models was assessed based on the correlation coefficient approaching ∼1 as well as low error values. Results indicated that the ANN model outperformed the RSM model, characterized by higher R<sup>2</sup> values and lower mean squared error (MSE) values, thereby suggesting superior predictive capability. Nevertheless, RSM effectively predicted the interaction among key process parameters and established their statistical significance as confirmed by analysis of variance (ANOVA) (p &lt; 0.05). These findings highlight the significance of both modeling approaches in process optimization and interpretation.</p> Graphical Abstract <p></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

Optimization and Modeling of Adsorptive Greywater Treatment System using RSM and ANN

  • 期刊论文

  • 发表日期：

    2025年11月26日

查看全文
  • Pushpraj Patel,
  • Shubhi Gupta,
  • Sanjeevani Hooda,
  • Prasenjit Mondal

摘要

The present study is backed by prior research from the authors (Patel et al. 2020), which further continued aiming to optimize and model the adsorption process for greywater treatment using adsorbents derived from pine needles, sugarcane bagasse, and sawdust. Initial analysis of greywater characteristics revealed concentrations of chemical oxygen demand (COD) and biochemical oxygen demand (BOD) as 554 and 120 mg/L, respectively. In batch adsorption experiments, sawdust activated carbon (SDAC) exhibited the highest efficiency, in removing 97.83% of COD and 95.83% of BOD, followed by sugarcane bagasse activated carbon (SCBAC) (91.85% of COD and 90% of BOD) and pine needle activated carbon (PNAC) (95.30% of COD and 93.33% of BOD), under operating condition (pH 7, dosage: 8 g/L, contact time: 240 min). An approach combining response surface methodology (RSM) with artificial neural network (ANN) was employed for the optimization of batch adsorption parameters and to address their individual limitations. The feasibility of the developed models was assessed based on the correlation coefficient approaching ∼1 as well as low error values. Results indicated that the ANN model outperformed the RSM model, characterized by higher R2 values and lower mean squared error (MSE) values, thereby suggesting superior predictive capability. Nevertheless, RSM effectively predicted the interaction among key process parameters and established their statistical significance as confirmed by analysis of variance (ANOVA) (p < 0.05). These findings highlight the significance of both modeling approaches in process optimization and interpretation.

Graphical Abstract