【作者】 胡一帆； 杨昌柱； 但锦锋； 濮文虹； 杨家宽；

【Author】 Hu Yifan ;Yang Changzhu;Dan Jinfeng;Pu Wenhong;Yang Jiakuan;Huazhong University of Science and Technology;

【机构】 华中科技大学；

【摘要】 本文利用人工神经网络构建了进水化学需氧量、水力停留时间、碱度、pH、挥发性有机酸浓度和氧化还原电位等运行条件对出水COD浓度的影响模型,并利用响应曲面法对神经网络的参数(隐含层神经元个数、初始μ值和初始权值与阈值)进行了优化。结果表明:优化的神经网络能灵活地应用于EGSB污水处理系统的模拟和运行控制等方面。更多还原

【Abstract】 In present study,the effects of operating parameters such as chemical oxygen demand in influent,hydraulic retention time, alkalinity, pH, volatile fatty acid concentration and oxidation-reduction potential on the effluent COD concentration of EGSB reactor were modeled by artificial neural network with parameters( the number of neurons in the hidden layer,initial μ value and initial weights and biases) optimized using response surface methodology. The neural network with an optimized topology turned out to be a feasible means to simulate and control the wastewater treatment by EGSB system.更多还原