Tarbiat Modares University, Civil Engineering Department, Environmental Engineering Division, Tehran, Iran.
J Hazard Mater. 2010 Jul 15;179(1-3):769-75. doi: 10.1016/j.jhazmat.2010.03.069. Epub 2010 Mar 23.
In this study, the results of 1-year efficiency forecasting using artificial neural networks (ANN) models of a moving bed biofilm reactor (MBBR) for a toxic and hard biodegradable aniline removal were investigated. The reactor was operated in an aerobic batch and continuous condition with 50% by volume which was filled with light expanded clay aggregate (LECA) as carrier. Efficiency evaluation of the reactors was obtained at different retention time (RT) of 8, 24, 48 and 72 h with an influent COD from 100 to 4000 mg/L. Exploratory data analysis was used to detect relationships between the data and dependent evaluated one. The appropriate architecture of the neural network models was determined using several steps of training and testing of the models. The ANN-based models were found to provide an efficient and a robust tool in predicting MBBR performance for treating aromatic amine compounds.
在这项研究中，使用人工神经网络 (ANN) 模型对移动床生物膜反应器 (MBBR) 进行了为期 1 年的效率预测，该模型用于去除有毒且难以生物降解的苯胺。反应器在有氧间歇和连续条件下运行，体积填充率为 50%，载体为轻质膨胀粘土骨料 (LECA)。在不同的停留时间 (RT) 8、24、48 和 72 小时下，进水 COD 从 100 到 4000mg/L 对反应器进行了效率评估。采用探索性数据分析方法来检测数据与依赖项之间的关系。通过对模型进行多次训练和测试，确定了神经网络模型的适当结构。基于 ANN 的模型被发现是一种有效的、稳健的工具，可用于预测处理芳香胺化合物的 MBBR 性能。
