基于灰色-神经网络的充填体强度预测  

Backfill strength prediction based on the grey-neural network

作　　者：张钦礼[1] 王艳利[1] 曹小刚[1] 

机构地区：[1]中南大学资源与安全工程学院,湖南长沙410083

出　　处：《化工矿物与加工》2011年第12期26-28,37,共4页Industrial Minerals & Processing

摘　　要：为了解各因素对充填体强度的影响规律,准确预测其强度,在综合分析影响充填体料浆质量的各种因素的基础上,基于灰色关联度理论,确定人工神经网络的输入参数,建立充填体强度预测的神经网络模型。用试验所得数据作为训练样本和测试样本,将经过网络预测的结果与实际值进行对比分析。研究表明该灰色-神经网络模型预测精度高,可以满足充填体强度预测的要求。In order to understand the influence law of each factors on backfilling strength and to predict accurately the strength,the inputed parameters of Artificial Neural Network were detemined and the neural network mode of backfill strength prediction was establish based on the comprehensive analysis on the backfill slurry quality factors and grey correlation.The mode prediction results was compared with the factual values by using the experiment data as the training and test sample,which shows the grey-neural network mode has a higher prediction precision and can satisfy the backfill strength prediction.

关 键 词：充填体强度 灰色关联度分析 神经网络 预测模型 

分 类 号：TD853.34[矿业工程—金属矿开采]