基于GA-BP神经网络的金精矿品位的预测

在对金矿生产过程进行大量实际调研工作的基础上,分别采用BP神经网络和遗传算法优化BP神经网络的方法,建立了金精矿品位的预测模型,以现场采集的978组数据作为样本,运用噪声平滑技术进行数据预处理,筛选了770组数据,运用其中的650组数据建模,并运用其余的120组数据对模型进行了验证.通过对两个模型的预测误差分析,得出用遗传算法优化的BP神经网络(GA-BP)预测精度更高,当预测相对误差在±2%范围内时,模型的预测精度达到97.5%.

Concentrate Grade Prediction of Gold Ore Based on GA-BP Neural Network

Two prediction models for concentrate grade of gold mine were established respectively by using BP neural network and GA-BP neural network method on the basis of investigation in actual production. 978 groups of data were gathered from actual production， from which the 770 groups was selected for establishing the models， among which 120 groups was used for verification. By analyzing the predictive errors of two models， it is approved that the prediction model based on GA-BP neural network can provide better accuracy: when the relative prediction errors are within ±2%， the prediction accuracy reaches 97.5%.