Oilsand slurry image and video analysis

点击翻译按钮获取中文摘要

In oil sands mining, mined ore which is rich in bitumen is crushed, screened, mixed with water, and transported in a pipeline before extraction takes place to separate bitumen from sand. Recovery rate of bitumen fluctuates and it is known that a portion of the bitumen can be lost to tails due to suboptimal conditions of the extraction process. We propose to conduct collaborative research with Syncrude Canada - the largest oil sands producer of the world - to gain a further scientific understanding of the recovery process by using image processing techniques and data analysis. This understanding will help to improve bitumen recovery rate, and result in benefits that are both economical - through increased production - and environmental - through reduced bitumen in tails.Specifically, we will develop image processing algorithms to study images of the moving suspended particles or droplets in ore slurry sampled by a novel sensor developed recently by Syncrude, called slurry image analyzer (SIA). These real-time algorithms will derive the characteristics of droplets of both bitumen and sand such as their size, direction and magnitude of motion, shape, and color. Since these characteristics directly translate into bitumen recovery rate, when they are examined under different extraction conditions that can be controlled, Syncrude can then attempt to construct a model that relates the conditions to the recovery rate. With the help of this model, Syncrude's SIA can be installed online to allow field engineers to monitor recovery rate, adjust operating conditions, and achieve optimal extraction performance.