Effect of line pressure on inspection tool velocity
By Dr Aidan O’Donoghue, Pipeline Research Ltd
Introduction
A newly installed onshore pipeline was required to be inspected within the first 2 years of operation. Initially the line pressure was expected to be low (20 barg outlet) but after two years the pressure would increase to 50 bar with higher flow. It was essential to understand the pig velocity profile for both these cases.
Pipeline Research Ltd's
PIGLAB Compressible is an inhouse
simulation program used to investigate the motion of pigs in gas pipelines.
Setup
The pipeline is 340 km long and involves a hilly terrain with multiple pipeline wall thickness changes due to river and road crossings. Typical pig differential pressures for the various wall thicknesses were measured in a trial at the inspection vendor. Other information such as the weight of the tool was also provided. The target was to keep the inspection pig velocity within 1 to 4 m/s (“In-specification” range).
Sensitivity analysis
The gas properties and operating data for the two cases were provided by the operator. This included the inlet flow and temperature for the year 1 case (20 barg outlet pressure) and the year 2 case (50 barg
outlet pressure). The analysis aimed to determine when the pigs were less than 1 m/s and greater than 4 m/s (“Out-of-specification”).
Output
The pig velocity profile for the low and high pressures cases was calculated using PIGLAB Compressible: -
The pig travels at out-of-specification velocity for a total of 15 km in the low pressure case (blue plot) and only 1.2 km in the high pressure case (red plot). At high pressure, the tool is never above 4 m/s.
Commentary
The analysis allowed the client to understand the percentage of the line that would be within velocity specification for the inspection and to determine the optimum timing for the pipeline base line survey. Further sensitivities were performed using the PIGLAB Compressible model to examine temperature changes, flow disruptions and contingency measures.
<<< Previous Next >>>
Also in this issue: