膜科学与技术

人工神经网络在量化膜污染界面作用力中的应用

作者：陈镒锋，申利国，林红军

单位：浙江师范大学地理与环境科学学院，金华 321004

关键词：人工神经网络；膜生物反应器；膜污染；界面作用力

出版年,卷(期):页码： 2020,40(3):47-54

摘要：

膜生物反应器（MBR）中膜污染的粘附过程由污染颗粒和粗糙膜表面之间的界面作用力决定。高级extended-Derjaguin-Landau-Verwey-Overbeek（XDLVO）方法无法快速实现界面作用力的量化。本研究提出了径向基函数（RBF）人工神经网络（ANN）、反向传播（BP）ANN和广义回归神经网络（GRNN）三种模型用于量化这些界面作用力。RBF ANN，BP ANN和GRNN的预测结果均具有较高的回归系数和准确性，表明它们能很好地捕捉界面作用力与各种因素之间复杂的非线性映射关系。与高级XDLVO方法相比，RBF ANN，BP ANN和GRNN的量化效率均得到了显著提高。同时，BP ANN的预测性能优于RBF ANN和GRNN模型。案例研究进一步证明了BP ANN用于界面作用力定量分析的可行性。本研究为量化与膜污染相关的界面作用力提供了新方法。

Interfacial energy between sludge foulants and rough membrane surface critically determines adhesive fouling in membrane bioreactors (MBRs). As a current available method, the advanced extended Derjaguin-Landau-Verwey-Overbeek (XDLVO) approach cannot efficiently quantify the interfacial energy. In this study, novel methods including radial basis function (RBF) artificial neural network (ANN), back propagation (BP) ANN and generalized regression neural network (GRNN) were proposed to quantify the interfacial energy associated with the membrane fouling in an MBR. The prediction results of RBF ANN, BP ANN and GRNN have high regression coefficients and accuracies, suggesting their high capacity to capture the complicated non-linear mapping relations between interfacial energy and various factors. As compared with the advanced XDLVO approach, both RBF ANN, BP ANN and GRNN showed remarkably improved quantification efficiency. Meanwhile, BP ANN showed better prediction performance than RBF ANN and GRNN model. Case study further demonstrated the robustness and feasibility of BP ANN for interfacial energy quantification. This study provided a new approach to quantify interfacial energy associated with membrane fouling.

第一作者简介：陈镒锋（1995－），男，浙江海宁人，硕士研究生，主要研究方向为膜污染控制．E-mail：yfchen@zjnu.edu.cn *通讯作者，E-mail：hjlin@zjnu.cn

参考文献：

[1] Meng F, Zhang S, Oh Y, et al. Fouling in membrane bioreactors: An updated review [J]. Water Res., 2017, 114: 151-180.
[2] Aslam M, Charfi A, Lesage G, et al. Membrane Bioreactors for Wastewater Treatment: A review of mechanical cleaning by scouring agents to control membrane fouling [J]. Chem. Eng. J., 2017, 307: 897-913.
[3] Hoek E M V, Agarwal G K. Extended DLVO interactions between spherical particles and rough surfaces [J]. J. Colloid Interface Sci., 2006, 298(1): 50-58.
[4] Teng J, Shen L, He Y, et al. Novel insights into membrane fouling in a membrane bioreactor: Elucidating interfacial interactions with real membrane surface [J]. Chemosphere, 2018, 210: 769-778.
[5] Cai X, Zhang M, Yang L, et al. Quantification of interfacial interactions between a rough sludge floc and membrane surface in a membrane bioreactor [J]. J. Colloid Interface Sci., 2017, 490: 710-718.
[6] Van Oss C J. Hydrophobicity of biosurfaces — Origin, quantitative determination and interaction energies [J]. Colloids Surf., B, 1995, 5(3): 91-110.
[7] 寇朝卫, 张干伟, 沈舒苏, 等. 基于XDLVO理论解析膜法水处理过程中膜污染问题的研究 [J]. 膜科学与技术, 2017, 37(01): 8-15.
[8] Teng J, Zhang M, Leung K-T, et al. A unified thermodynamic mechanism underlying fouling behaviors of soluble microbial products (SMPs) in a membrane bioreactor [J]. Water Res., 2019, 149: 477-487.
[9] Singh K P, Gupta S. Artificial intelligence based modeling for predicting the disinfection by-products in water [J]. Chemom.?Intell.?Lab.?Syst., 2012, 114: 122-131.
[10] Ghritlahre H K, Prasad R K. Investigation of thermal performance of unidirectional flow porous bed solar air heater using MLP, GRNN, and RBF models of ANN technique [J]. Therm. Sci. Eng. Prog., 2018, 6: 226-235.
[11] Iliyas S A, Elshafei M, Habib M A, et al. RBF neural network inferential sensor for process emission monitoring [J]. Control Eng. Pract., 2013, 21(7): 962-970.
[12] Chen Y, Yu G, Long Y, et al. Application of radial basis function artificial neural network to quantify interfacial energies related to membrane fouling in a membrane bioreactor [J]. Bioresour. Technol., 2019, 293: 122103.
[13] Van Oss C J. Acid—base interfacial interactions in aqueous media [J]. Colloids Surf. A, 1993, 78: 1-49.
[14] Meinders J M, Van Der Mei H C, Busscher H J. Deposition Efficiency and Reversibility of Bacterial Adhesion under Flow [J]. J. Colloid Interface Sci., 1995, 176(2): 329-341.
[15] Brant J A, Childress A E. Colloidal adhesion to hydrophilic membrane surfaces [J]. J. Membr. Sci., 2004, 241(2): 235-248.
[16] Adam N K. Use of the Term ‘Young's Equation’ for Contact Angles [J]. Nature, 1957, 180: 809.
[17] Bhattacharjee S, Elimelech M. Surface Element Integration: A Novel Technique for Evaluation of DLVO Interaction between a Particle and a Flat Plate [J]. J. Colloid Interface Sci., 1997, 193(2): 273-285.
[18] Bhattacharjee S, Ko C-H, Elimelech M. DLVO interaction between rough surfaces [J]. Langmuir, 1998, 14(12): 3365-3375.
[19] Shetty G R, Chellam S. Predicting membrane fouling during municipal drinking water nanofiltration using artificial neural networks [J]. J. Membr. Sci., 2003, 217(1): 69-86.
[20] Huang G B, Chen L, Siew C K. Universal approximation using incremental constructive feedforward networks with random hidden nodes [J]. IEEE Trans. Neural Netw., 2006, 17(4): 879-892.
[21] Dreiseitl S, Ohno-Machado L. Logistic regression and artificial neural network classification models: a methodology review [J]. J. Biomed. Inform., 2002, 35(5-6): 352-359.
[22] Ghritlahre H K, Prasad R K. Application of ANN technique to predict the performance of solar collector systems - A review [J]. Renewable Sustainable Energy Rev., 2018, 84: 75-88.

服务与反馈：

【文章下载】【加入收藏】

《膜科学与技术》编辑部地址：北京市朝阳区北三环东路19号蓝星大厦邮政编码：100029 电话：010-64426130/64433466 传真：010-80485372邮箱：mkxyjs@163.com