Diagnosis of concentrate grade and mass flowrate in tin flotation from colour and surface texture analysis
Diagnosis of concentrate grade and mass flowrate in tin flotation from colour and surface texture analysis
复制标题
DOI：
10.1016/s0892-6875(97)00040-x
复制
发表时间：
1997-06-01
影响因子：
4.8
通讯作者：
Hall, ST
中科院分区：
文献类型：
--
作者：
Hargrave, JM;Hall, ST
长期以来，人们一直认为泡沫的视觉外观通常是浮选过程性能的一个很好的定性指标。本文描述了为研究利用这一前提开发一种利用图像分析的浮选过程在线控制系统的可能性而进行的工作。将图像分析和神经网络等新兴技术应用于某工业锡浮选厂浮选泡沫图像的研究。对图像进行处理，生成泡沫表面的颜色轮廓和纹理描述符。然后，利用传统的统计和神经网络模型，将这些数据与诸如精矿品位、水和固体流量等性能特征相关联。结果表明，泡沫的颜色可以用来预测精矿的品位。研究了描述泡沫表面结构的不同方法，包括一种新的分形分析方法。结果表明，泡沫的结构是浮选池性能的良好指标。预先讨论了在鲁棒控制系统中使用这种方法的可能性。1997爱思唯尔科学有限公司
It has long been accepted that the visual appearance of a froth is generally a good qualitative indicator as to the performance of the flotation process. This paper describes work which has been undertaken to investigate the possibility of using this premise to develop an on-line control system for the flotation process using image analysis.The emerging technologies of image analysis and neural networks have been applied to images of flotation froths captured at an industrial tin flotation plant. The images were processed to produce a colour profile and textural descriptors of the froth surface. This data was then correlated against performance characteristics, such as concentrate grade, water and solids flowrates, using both conventional statistical and neural network modelling.It is shown that the colour of the froth can be used to give a prediction of the grade of the concentrate produced. Different procedures of describing the surface texture of the froth were studied, including a novel method using fractal analysis. it is shown that the texture of the froth is a good indicator to the performance of the flotation cells. The possibilities of using this in a robust control system pre discussed. (C) 1997 Elsevier Science Ltd.