基于BP神经网络的氧化铝溶出苛性比值预测方法

发布时间：2018-03-15 17:25

本文选题：氧化铝　切入点：溶出苛性比值　出处：《轻金属》2017年09期 　论文类型：期刊论文

【摘要】：针对拜耳法生产氧化铝的溶出工序中苛性比值难以实时获得的问题,提出了一种通过BP神经网络模型对溶出苛性比值进行提前预测的方法。根据溶出生产物料输入输出关系,提出了基于物料平衡的溶出苛性比值机理计算模型,并对计算模型的输入变量进行优化,使其能够满足BP神经网络输入的要求。最终设计了一种具有误差反传学习及历史数据训练功能的BP神经网络,经山西某铝厂实际数据测试,BP神经网络能够较好的预测溶出苛性比值。
[Abstract]:In order to solve the problem that it is difficult to obtain the ratio of causticity in the digestion process of alumina production by Bayer process in real time, a BP neural network model is proposed to predict the ratio of dissolution causticity in advance. According to the relation between input and output of dissolved materials, A model for calculating dissolution causticity ratio based on material balance is proposed, and the input variables of the model are optimized. Finally, a BP neural network with the function of error backpropagation learning and historical data training is designed. The BP neural network can predict the ratio of dissolution causticity better by the actual data of an aluminum factory in Shanxi province.
【作者单位】：沈阳铝镁设计研究院有限公司;
【分类号】：TP183;TQ133.1

【相似文献】