利用人工神经网络建立水滑石晶化过程的关联方程  

Neural Networks for Study of Mg-Al Hydrotalcite of Continus Aging Process

作　　者：黄小葳[1] 李殿卿[2] 段雪[2] 

机构地区：[1]北京联合大学生物化学工程学院,北京100023 [2]北京化工大学可控化学反应科学与技术教育部重点实验室,北京100029

出　　处：《化工时刊》2006年第2期12-15,共4页Chemical Industry Times

摘　　要：采用人工神经网络方法对镁铝水滑石的连续晶化过程进行了研究，以便找出最佳的晶化条件。本文利用神经网络建立的模型，分析了成核料液的浓度、流量以及晶化温度、晶化时间对形成的水滑石晶体的影响，并由此建立了关联方程。研究表明最佳晶化条件：浓度在0．4～0．6mol/mL，流量在18mL／min左右，晶化时间400min左右，晶化温度100℃。Artificial neural network （ANN）were used to stuky continous aging process of Mg- Al hyduotalcite to seek optimum operating condition. The influences of the operating condition of continue aging process such as concentuation, folw, temperature and aning time for hydrotalcite latticewas analyzed. On Basis of experimental data, the neural network was build and the model of relation were got. The Study show that concentration are in 0.4-0.6mol/ml, the flow are in 18 mL/min and aging time in 100℃ under the experimental conditions.

关 键 词：人工神经网绍 镁铝水滑石 连续晶化过程 晶胞参数 

分 类 号：TQ026[化学工程]