Leak detection that manages the changing densities of slurry and thickened tailings in pipelines

Long distance slurry pipelines are widely used in the mining industry and, compared to bulk transport systems, have a good performance record for high-availability, excellent safety, and profitability in transporting mineral concentrates such as coal, copper, iron, phosphates, and oil sands. Mining operations also use shorter pipelines to pump tailings or water. The abrasive products transported can erode the pipeline wall, causing a leak or even a rupture. Pipelines in areas prone to mudslides or earthquakes are at even greater risk of a spill, and such occurrences are far from rare in South America.

The non-Newtonian nature of slurry make model-based less effective for leak detection

Unlike the model-based leak detection systems historically used on slurry pipelines, the Atmos Pipe statistical volume balance system does not attempt to model the composition and viscosity of the slurry or tailings transported in a pipeline. Atmos Pipe thus avoids the high false alarm rate and poor accuracy displayed by real-time transient models because real-time transient models cannot accurately model the continually changing composition and viscosity of mineral compounds in a slurry pipeline.

Ore particles can fall out of suspension following an unplanned shutdown and block the line

Leak detection in slurry pipelines must minimize false leak alarms because stopping the flow could block a pipeline as the particles fall out of suspension, causing expensive downtime. Sometimes it is even necessary to cut the pipeline to clear the blockage. Each shutdown can cost hundreds of thousands in lost revenue. The paramount importance of high-reliability was the perfect opening for the statistical volume balance leak detection to displace model-based leak detection systems when it entered the mining space over 15 years ago. Since then, its supreme reliability has earned the trust of pipeline controllers on some of the most complex slurry pipelines in the mines of South America.

The nature of slurry products makes accurate and repeatable flow measurement more challenging than other products. Atmos Pipe uses patented algorithms and self-trained filters to automatically compensate for those measurement errors to maximize sensitivity and reliability of leak detection performance. This is another reason the statistical leak detection system has proven far more reliable than model-based leak detection systems on slurry pipelines over the past 15 years, significantly reducing the number of shut-downs and saving the pipeline operators thousands of dollars.

Categories: Best practice advice

By: Atmos International
Date: 12 April 2019