Artificial neural network model for extracting knowledge from the electro-Fenton process for acid mine wastewater treatment

IF 1.5 4区环境科学与生态学 Q4 ENVIRONMENTAL SCIENCES

Clean-soil Air Water Pub Date : 2024-08-30 DOI:10.1002/clen.202400029

Anoop Kumar Maurya, Pasupuleti Lakshmi Narayana, Uma Maheshwera Reddy Paturi, Subba Reddy Nagireddy Gari

{"title":"Artificial neural network model for extracting knowledge from the electro-Fenton process for acid mine wastewater treatment","authors":"Anoop Kumar Maurya, Pasupuleti Lakshmi Narayana, Uma Maheshwera Reddy Paturi, Subba Reddy Nagireddy Gari","doi":"10.1002/clen.202400029","DOIUrl":null,"url":null,"abstract":"In this study, artificial neural networks (ANNs) were employed to analyze the complex interactions between electro-Fenton (EF) process variables (plate spacing, current intensity [CI], initial pH, aeration rate) and the Fe(II) and Mn(II) removal efficiency from wastewater. After experimenting with 69 different ANN architectures, the 4-8-8-2 architecture was identified as more efficient, achieving higher accuracy (adj. R2 of 0.93 for Fe(II) and 0.96 for Mn(II)) than the published model. The research provides valuable insights into the correlation between EF process parameters and removal efficiency, guiding the optimization of wastewater treatment processes. Sensitivity analysis revealed that CI significantly affects Mn(II) and Fe(II) removal efficiency. A user-friendly graphical interface was created based on the synaptic weights of the best model to enable practical predictions. It is designed to be accessible even to users without programing experience.","PeriodicalId":10306,"journal":{"name":"Clean-soil Air Water","volume":"52 10","pages":""},"PeriodicalIF":1.5000,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Clean-soil Air Water","FirstCategoryId":"93","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/clen.202400029","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}

引用次数: 0

Abstract

In this study, artificial neural networks (ANNs) were employed to analyze the complex interactions between electro-Fenton (EF) process variables (plate spacing, current intensity [CI], initial pH, aeration rate) and the Fe(II) and Mn(II) removal efficiency from wastewater. After experimenting with 69 different ANN architectures, the 4-8-8-2 architecture was identified as more efficient, achieving higher accuracy (adj. R² of 0.93 for Fe(II) and 0.96 for Mn(II)) than the published model. The research provides valuable insights into the correlation between EF process parameters and removal efficiency, guiding the optimization of wastewater treatment processes. Sensitivity analysis revealed that CI significantly affects Mn(II) and Fe(II) removal efficiency. A user-friendly graphical interface was created based on the synaptic weights of the best model to enable practical predictions. It is designed to be accessible even to users without programing experience.

Abstract Image

查看原文本刊更多论文

从用于酸性矿山废水处理的电-芬顿工艺中提取知识的人工神经网络模型

本研究采用人工神经网络（ANN）分析了电-芬顿（EF）工艺变量（板间距、电流强度 [CI]、初始 pH 值、曝气速率）与废水中铁（II）和锰（II）去除率之间复杂的相互作用。在尝试了 69 种不同的 ANN 架构后，4-8-8-2 架构被认为更有效，比已发表的模型具有更高的准确性（铁（II）的 R2 值为 0.93，锰（II）的 R2 值为 0.96）。这项研究为了解 EF 工艺参数与去除效率之间的相关性提供了宝贵的见解，为优化废水处理工艺提供了指导。敏感性分析表明，CI 对锰（II）和铁（II）的去除效率有显著影响。根据最佳模型的突触权重创建了一个用户友好型图形界面，以便进行实际预测。即使没有编程经验的用户也可以使用该界面。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Clean-soil Air Water 环境科学-海洋与淡水生物学

CiteScore

2.80

自引率

5.90%

发文量

审稿时长

3.6 months

期刊介绍： CLEAN covers all aspects of Sustainability and Environmental Safety. The journal focuses on organ/human--environment interactions giving interdisciplinary insights on a broad range of topics including air pollution, waste management, the water cycle, and environmental conservation. With a 2019 Journal Impact Factor of 1.603 (Journal Citation Reports (Clarivate Analytics, 2020), the journal publishes an attractive mixture of peer-reviewed scientific reviews, research papers, and short communications. Papers dealing with environmental sustainability issues from such fields as agriculture, biological sciences, energy, food sciences, geography, geology, meteorology, nutrition, soil and water sciences, etc., are welcome.