By-pass pigging

Abstract

Pipelines are used in many industries as a means of transporting fluids, for example in the oil and gas industry. Pigs are devices that move through such pipelines, for instaoce to cleao the pipeline or to perform internal inspections. They are driven by a pressure difference over the pigs. Since this is a risky operation, there is a strong motivation to control the motion of these pigs. One possibility is to use so-called by-pass pigs. These pigs have a hole through their body such that fluids cao by-pass the device. This lowers the pig velocity. If the by-pass area cao be varied during a pigging operation, it is possible to control the pig velocity. This concept is relatively new aod not yet completely understood. Research is currently carried out at the TU Delft in collaboration with Shell to get abetter understaoding of the behaviour of conventional pigs of by-pass pigs. This MSc thesis is part of that project aod focuses on performing experiroents for pigging operations in a laboratory environment. The relevance is two-fold From one haod, more insight will be obtained in the encountered phenomena From the other hand, the results can be used to validate the numerical pigging model which is currently in development. The experiroents were carried out in a flowloop at the department of Process & Energy at the TU Delft. The flowloop has a length of 65 m aod a diameter of about 52 mm. Air is used as working fluid aod the flowloop has an atroospheric outlet pressure. Duriog the pigging experiroents, the bulk velocity aod pressure in the flowloop were recorded. Three cameras were used for visual observation from which the velocity was deduced. The modular pig design made it possible to quickly chaoge between different pig configurations. It turned out that small variations in this configuration can have large influence on the pig motion. A pre­ dominant characteristic of the pig motion is the so-called stick-slip motion. This motion is characterized by a quick acceleration and deceleration of the pig as a consequence of a varying friction. A module was added to the numerical pigging model to include the effect of variations in the friction, which forces the siroula­ tion to give a stick-slip behaviour. Besides this, also an aoalytical approach was taken to obtain first insights into this behaviour. The experiroental results show that the maximum pig velocity cao become significantly larger thao the average pig velocity. The ratio of the maximum velocity over the average velocity increases at lower bulk velocities. The stick-slip models cao give a reasonable good estimation of the maximum pig velocity. Besides this, they predict a similar trend in the pressure fluctuations. To compare the influence of the bulk velocity aod the by-pass area, ao extensive parameter study was carried out. Results of 132 pigging runs with two different types of sealing were included. The by-pass area was varied from O % to 4 % and the bulk velocity was varied in the raoge of 1.5 ml s to 7 ml s . The focus was on the effects of these chaoges on the average pig velocity, the average friction and the staodard deviation of the pressure. It turned out that the friction of the pigs used in the experiroents was not depending on the velocity. The staodard deviation of the pressure, which is a measure of the intensity of the stick-slip behaviour, was different for both types of sealing. However, for a certain configuration no dependence on the pig velocity was found. The average pig velocityitselfis largely dependent on the by-pass area. The results were compared with the numerical pigging model, a commercial package and a steady-state analytical model. Itturned out that the average velocity cao accurately be determined with all these models, even if the pig motion shows a strong stick-slip behaviour.