Evaluation of artificial neural network algorithms for predicting the effect of the urine flow rate on the power performance of microbial fuel cells.
Evaluation of artificial neural network algorithms for predicting the effect of the urine flow rate on the power performance of microbial fuel cells.
复制标题
DOI：
10.1016/j.energy.2020.118806
复制
发表时间：
2020-12-15
期刊：
影响因子：
--
通讯作者：
Ieropoulos IA
中科院分区：
文献类型：
--
作者：
de Ramón-Fernández A;Salar-García MJ;Ruiz Fernández D;Greenman J;Ieropoulos IA
微生物燃料电池（MFC）的功率性能很大程度上取决于生物膜的生长，而生物膜的生长又受到进料流量的影响。在这项工作中，采用人工神经网络 (ANN) 方法来模拟流速对用纯人类尿液喂养的陶瓷 MFC 功率输出的影响。为此，使用三种不同的二阶算法来训练我们的网络，然后在预测精度和收敛时间方面进行比较：拟牛顿算法、莱文伯格-马夸特算法和共轭梯度算法。结果表明，三种训练算法能够准确模拟发电量。其中，Levenberg-Marquardt 的精度最高（R = 95%），收敛速度最快（7.8 秒）。这些结果表明，ANN 是预测可变流量条件下陶瓷 MFC 能量收集的有用且可靠的工具，这将有助于该技术的实际部署。人工神经网络是预测陶瓷 MFC 功率性能的可靠工具。 QN、LM 和 CG 算法能够准确模拟 MFC 的功率输出。 LM 算法显示出最高的精度（R = 95%）和最快的收敛速度（7.8s）。
Microbial fuel cell (MFC) power performance strongly depends on the biofilm growth, which in turn is affected by the feed flow rate. In this work, an artificial neural network (ANN) approach has been used to simulate the effect of the flow rate on the power output by ceramic MFCs fed with neat human urine. To this aim, three different second-order algorithms were used to train our network and then compared in terms of prediction accuracy and convergence time: Quasi-Newton, Levenberg-Marquardt, and Conjugate Gradient. The results showed that the three training algorithms were able to accurately simulate power production. Amongst all of them, the Levenberg-Marquardt was the one that presented the highest accuracy (R = 95%) and the fastest convergence (7.8 s). These results show that ANNs are useful and reliable tools for predicting energy harvesting from ceramic-MFCs under changeable flow rate conditions, which will facilitate the practical deployment of this technology. ANNs are reliable tools for predicting the power performance of ceramic-MFCs. QN, LM and CG algorithms were able to accurately simulate the power output by MFCs. LM algorithm showed the highest accuracy (R = 95%) and the fastest convergence (7.8s).
登录
查看更多内容
DOI：
10.1016/j.energy.2017.11.135
发表时间：
2018-02-01
期刊：
Energy (Oxford, England)
影响因子：
--
作者：
Gajda I;Greenman J;Santoro C;Serov A;Melhuish C;Atanassov P;Ieropoulos IA
通讯作者：
Ieropoulos IA
影响因子：
9.2
作者：
Garcia, Maria Jose Salar;Santoro, Carlo;Ieropoulos, Ioannis
通讯作者：
Ieropoulos, Ioannis
影响因子：
3.3
作者：
Ieropoulos, Ioannis;Greenman, John;Melhuish, Chris
通讯作者：
Melhuish, Chris
将微生物群落数据与机器学习技术相结合来预测微生物燃料电池中的进料底物
影响因子：
12.6
作者：
Cai, Wenfang;Lesnik, Keaton Larson;Liu, Hong
通讯作者：
Liu, Hong
影响因子：
64.8
作者：
RUMELHART, DE;HINTON, GE;WILLIAMS, RJ
通讯作者：
WILLIAMS, RJ