刘洋, 涂宁宇, 谢文玉, 金仁和, 李友明. 基于GMDH-type神经网络优化油页岩吸附铜离子的研究[J]. 华南师范大学学报(自然科学版), 2013, 45(4).
引用本文: 刘洋, 涂宁宇, 谢文玉, 金仁和, 李友明. 基于GMDH-type神经网络优化油页岩吸附铜离子的研究[J]. 华南师范大学学报(自然科学版), 2013, 45(4).
Optimization of Adsorption Process of Cu(Ⅱ) Ion by Oil Shale Base on GMDH-type neural network[J]. Journal of South China Normal University (Natural Science Edition), 2013, 45(4).
Citation: Optimization of Adsorption Process of Cu(Ⅱ) Ion by Oil Shale Base on GMDH-type neural network[J]. Journal of South China Normal University (Natural Science Edition), 2013, 45(4).

基于GMDH-type神经网络优化油页岩吸附铜离子的研究

Optimization of Adsorption Process of Cu(Ⅱ) Ion by Oil Shale Base on GMDH-type neural network

  • 摘要: 利用GMDH前馈型神经网络优化油页岩吸附金属铜离子实验,设定吸附质/吸附剂、pH、反应时间为自变量,吸附率为因变量,建立吸附数学模型对吸附过程进行预测.根据GMDH神经网络模型分析,发现pH对于吸附率的影响权重最大,同时很好的诠释了3种自变量条件对于吸附率的作用机理.此外,利用神经网络模型进行模拟实验,预测值拟合Langmuir吸附等温线,相关系数达到0.907.证明建立的神经网络数学模型与经典吸附理论吻合,且精度很高.

     

    Abstract: GMDH-type neural network was used to optimize the experiment of Cu(Ⅱ) ion absorption by oil shale. In order to build a mathematical adsorption model on the absorption process, mass proportion for absorbate and absorbent, pH value and contact time were considered as independent variables, while the absorption rate as dependent variable. According to the analysis by using the GMDH-type neural network model, the effect of pH values on the absorption rate is higher than the others. Additionally, GMDH-type neural network explains the absorption mechanism based on the three variables. What’s more, a comparative study was made to confirm the good correlation between GMDH and Langmuir models on sorption isotherms, while the coefficient of correlation was 0.907.

     

/

返回文章
返回