定性与定量信息相结合预测金品位的方法研究  

The Method of Combining Qualitative Information and Quantitative Information to Predict Gold Grade

作　　者：梁智霖 郭攀[2] LIANG Zhilin;GUO Pan(Basic Department of Henan Health Executive College,Zhengzhou 450000,China;School of Water Resources and Civil Engineering,Zhengzhou University,Zhengzhou 450000,China)

机构地区：[1]河南卫生健康干部学院,河南郑州450000 [2]郑州大学水利与交通学院,河南郑州450000

出　　处：《湿法冶金》2024年第2期195-200,共6页Hydrometallurgy of China

摘　　要：将改进云模型和改进RBF神经网络相结合,提出了一种预测矿石中金品位的模型。先利用DS证据理论和云模型将定性信息定量化,再采用量子粒子群算法和RBF神经网络完成矿石中金品位预测。结果表明:该模型的均方根误差为0.0092,最大误差为0.0161,相关系数为0.9402,可较好保留定性信息特性,金品位预测效果较好。A gold grade prediction model was proposed by combining improved cloud models with improved RBF neural networks.Qualitative information was quantified using DS evidence theory and cloud models,and then quantum particle swarm optimization algorithm and RBF neural network were used to predict the gold grade in ores.The results indicate that the mean square error of this model is 0.0092,the maximum error is 0.0161,and the correlation coefficient is 0.9402,the model can better preserve the qualitative information characteristics,the prediction effect of gold grade is good.

关 键 词：金 品位 预测 模型 定性信息 定量信息 云模型 RBF神经网络 

分 类 号：TF803.21[冶金工程—有色金属冶金]