NiFe_2O_4基金属陶瓷耐腐蚀因素分析及腐蚀率预测  

Corrosion Analysis and Corrosion Rates Prediction ofNiFe_2O_4 Cermet Inert Anodes

作　　者：赖延清[1] 陈湘涛[1] 秦庆伟[1] 李劼[1] 刘业翔[1] 

机构地区：[1]中南大学冶金科学与工程学院,湖南长沙410083

出　　处：《中南大学学报（自然科学版）》2004年第6期896-901,共6页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(50204014);国家重点基础研究发展规划项目(G1999064903);国家高技术研究发展计划项目(2001AA335013)

摘　　要：采用灰关联分析方法解析了铝电解5%Ni NiFe2O4基金属陶瓷惰性阳极的电解腐蚀率与电解参数的关系,建立了预测惰性阳极腐蚀率的人工神经网络模型。研究结果表明:灰关联分析是一种新的惰性阳极腐蚀数据处理方法;根据灰关联度的计算,在很多电解参数中找出了影响惰性阳极腐蚀率的主要因素,即Al2O3质量浓度、电解温度、分子比、面积比和电流密度等,并指出了各因素对电极腐蚀的影响程度;对NiFe2O4基金属陶瓷惰性阳极电解腐蚀率的预测结果与实测值吻合,表明利用所建立的神经网络模型能有效地预测惰性阳极腐蚀率。The corrosion rates of 5%Ni-NiFe_2O_4 inert anodes were investigated as a function of main operating parameters by grey relational analysis. The results show that grey relational analysis is a novel data process method and is used to rank the correlation extent of effect factors in the inert anode system. The analysis of practical data show that the prior factors causing corrosion are alumina concentration, bath temperature, cryolitic ratio, area ratio of cathode to anode, and current density in turn. Corrosion analyses are consistent with experimental results. An artificial neural network(ANN)-based solution methodology for modeling anode corrosion procesed from observed experimental values, and ANN model were developed by using the cited methodology for the prediction of the corrosion rate of inert anode in aluminium electrolysis. The experimental result shows that ANN model holds promise to be an effective and efficient tool for the construction of analytical models associated with corrosion process of inert anodes in cryolite-based melts.

关 键 词：铝电解 惰性阳极 腐蚀 灰关联分析 人工神经网络 

分 类 号：TF821[冶金工程—有色金属冶金]