当前位置: X-MOL 学术Miner. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
In-line characterisation of continuous phase conductivity in slurry flows using artificial intelligence tomography
Minerals Engineering ( IF 4.9 ) Pub Date : 2021-09-21 , DOI: 10.1016/j.mineng.2021.107203
Thomas D. Machin 1, 2 , Kent Wei 3 , Richard W. Greenwood 1 , Mark J.H. Simmons 1
Affiliation  

Electrical Impedance Tomography (EIT) can be applied to monitor a variety of mineral and chemical processes including: velocity measurements in drilling cuttings and hydrocyclone operations. Hydraulic conveying systems rely upon the knowledge of slurry density to ensure efficient transportation of the solids. Typically, density measurements exploit the attenuation of gamma ray photons which poses complex safety, operational and regulatory concerns with Electrical Impedance Tomography affording a non-nuclear alternative to traditional approaches. To optimise the accuracy of this non-nuclear density measurement, the electrical conductivity of the aqueous phase in a multi-component slurry, is required. Whilst conductivity probes are sufficiently accurate, there are often drawbacks and limitations due to installation restrictions, as it is difficult to separate aqueous and solid phases in real-time. Electrical Impedance Fingerprinting (EIF), is a novel measurement technique which characterises formulation properties, in-situ, based upon electrical impedance sensing and artificial intelligence algorithms. This paper outlines the development of EIF and its application to monitor aqueous phase conductivity in multi-component slurries, containing sands and clays. EIF accurately predicts this conductivity with high accuracy and a root-mean squared error of 0.055 mS cm−1. This development ensures accurate non-nuclear density measurements (<5%) are obtained across an extended aqueous electrical conductivity range of 1.5–70 mS cm−1. This encompasses the majority of target hydraulic conveying systems in mining operations. EIF also enhances the functionality of ‘traditional’ electrical tomography as not only are mineral processes able to be visualised, but the process materials are simultaneously characterised, to improve process understanding, optimisation and control.



中文翻译:

使用人工智能断层扫描在线表征浆料流中的连续相电导率

电阻抗断层扫描 (EIT) 可用于监测各种矿物和化学过程,包括:钻屑和水力旋流器操作中的速度测量。液压输送系统依靠泥浆密度的知识来确保固体的有效运输。通常,密度测量利用伽马射线光子的衰减,这会引起复杂的安全、操作和监管问题,而电阻抗断层扫描提供了传统方法的非核替代方案。为了优化这种非核密度测量的准确性,需要多组分浆液中水相的电导率。虽然电导率探头足够准确,但由于安装限制,通常存在缺陷和限制,因为很难实时分离水相和固相。电阻抗指纹 (EIF) 是一种新颖的测量技术,它基于电阻抗传感和人工智能算法,原位表征配方特性。本文概述了 EIF 的发展及其在含砂和粘土的多组分浆料中监测水相电导率的应用。EIF 以高精度和 0.055 mS cm 的均方根误差准确地预测了这种电导率 本文概述了 EIF 的发展及其在含砂和粘土的多组分浆料中监测水相电导率的应用。EIF 以高精度和 0.055 mS cm 的均方根误差准确地预测了这种电导率 本文概述了 EIF 的发展及其在含砂和粘土的多组分浆料中监测水相电导率的应用。EIF 以高精度和 0.055 mS cm 的均方根误差准确地预测了这种电导率-1。这一发展确保在 1.5–70 mS cm -1的扩展水性电导率范围内获得准确的非核密度测量 (<5%) 。这包括采矿作业中的大多数目标液压输送系统。EIF 还增强了“传统”电子断层扫描的功能,因为不仅可以可视化矿物过程,而且可以同时表征过程材料,以提高过程理解、优化和控制。

更新日期:2021-09-21
down
wechat
bug