基于GA-BP神经网络的浮选加药量预测

摘要

针对现有煤泥浮选加药量预测不准确的问题,提出了基于GA-BP神经网络作为煤泥浮选加药量的预测模型.首先通过MIV值评价法筛选出对浮选加药量影响较大的因素,进而建立了基于GA-BP神经网络的加药预测模型.用MIV值评价法完成了对网络结构的简化;用遗传算法优化神经网络的方法提高了神经网络模型预测加药量的精度.%The prediction model of coal slime flotation dosage based on GA-BP neural network is presented, aiming at the poor accuracy of current coal slime flotation dosage prediction. Firstly, the influence factors were screened out through MIV value method, and then the prediction model based on GA-BP neural network of dosing was established. MIV value method was used to simplify the network structure, and the method of genetic algorithm was employed to optimize the neural network and improve the accuracy of the neural network model to predict the dosage.