硫化矿石常温氧化模拟及基于神经网络的氧化活性预测

来源期刊:中南大学学报(自然科学版)2020年第4期

论文作者:潘伟 刘正洲 吴爱祥 吴超 邓舵 冯宴熙 杨奉龙

文章页码:863 - 872

关键词:硫化矿石;常温氧化;质量增大率;预测模型;神经网络

Key words:sulfide ores; normal temperature oxidation; mass increase rate; prediction model; neural network

摘    要:在实验室内开展硫化矿石常温氧化实验。以样品粒度、初始含水率和预氧化时间这3个因素作为输入单元,质量增大率作为输出单元,建立样品质量增大率的神经网络预测模型。研究结果表明:硫化矿石在常温条件下质量变化趋势包括3个阶段,依次为快速增大阶段、增幅减小阶段和保持不变阶段;未被氧化的矿样表面比较光洁,粒度分布较均匀,氧化后的矿样表面有明显结块现象;样品质量增大率与粒度和环境pH均呈负相关关系;随着初始含水率和预氧化时间增大,样品质量增大率均呈先增大后减小趋势;高温高湿环境可促进硫化矿石氧化;3个因素对样品质量增大率的影响重要度从高至低依次为初始含水率、预氧化时间和粒度;建立的神经网络模型具有较高的预测精度,相对误差小于10 %,可用于对实测样品质量增大率的可靠性进行验证。

Abstract: Normal temperature oxidation experiments of sulfide ores were carried out in laboratory. A neural network prediction model of mass increase rate of samples was established by taking particle size, initial moisture content and pre-oxidation time as input units and mass increase rate of samples as output units. The results show that mass trend of sulfide ore samples under normal temperature conditions includes three stages, i.e., rapidly increasing stage, slowly increasing stage and basically unchanged stage. Before oxidation, the surface of ore samples is relatively smooth and the particles are relatively uniform. After oxidation, there is obvious agglomeration phenomenon on the surface of ore samples. With the increase of particle size and environmental pH, mass increase rate of samples decreases. With the increase of initial moisture content and pre-oxidation time, mass increase rate of samples increases first and then decreases. High temperature and humidity can promote the oxidation of sulfide ores. Influence of the factors on mass increase rate of samples from high to low is initially moisture content, pre-oxidation time and ore particle size. The neural network has higher prediction accuracy, and the relative error is less than 10%. Therefore, it can be used to validate the reliability of the measured mass increase rate.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号