全尾砂浆最佳絮凝沉降参数

发布时间：2018-05-23 07:57

本文选题：絮凝剂单耗 + 尾砂　；参考：《中南大学学报(自然科学版)》2016年05期

【摘要】：为了得到最佳的絮凝沉降参数,研究使用BP神经网络进行优化选择。通过对比分析,将输入因子简化为絮凝剂单耗和尾砂质量分数2个因子,输出因子简化为沉降速度1个因子;通过正交试验,建立网络学习、训练样本,优选出最佳网络模型。扩大正交试验,增加输入因子水平,组合优选样本,搜索最佳絮凝沉降参数。以司家营铁矿全尾砂絮凝沉降为例,优选出絮凝剂单耗为10 g/t,尾砂质量分数为18%,预测沉降速度为1.38 m/h,满足生产要求,比原生产所需絮凝剂单耗节省50%。应用结果表明:该研究成果效果显著,为絮凝沉降参数优选提供一种新思路。
[Abstract]:In order to obtain the best flocculation settlement parameters, BP neural network was used to optimize selection. Through comparative analysis, the input factor is simplified as two factors of flocculant unit consumption and tailing mass fraction, the output factor is simplified as one factor of settling speed, and the best network model is selected by establishing network learning, training sample and optimizing network model through orthogonal experiment. Expand the orthogonal test, increase the level of input factor, select the best sample and search for the best flocculation settling parameters. Taking the flocculation settlement of total tailings in Sijiaying Iron Mine as an example, the unit consumption of flocculant is 10 g / t, the mass fraction of tailings is 18, and the predicted settling velocity is 1.38 m / h, which meets the production requirements and saves 50% of the unit consumption of flocculant compared with the original production. The application results show that the effect of this study is remarkable, which provides a new idea for the optimization of flocculation settlement parameters.
【作者单位】：中南大学资源与安全工程学院;
【基金】：科技部“十一五”科技支撑计划项目(2008BAB32B03)~~
【分类号】：TD926.4

【相似文献】