SciELO - Scientific Electronic Library Online

 
vol.31 issue2Electrodermal activity - a review author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Ingeniería e Investigación

Print version ISSN 0120-5609

Ing. Investig. vol.31 no.2 Bogotá May/Aug. 2011

 

Evaluating drag force and geometric optimisation of pipeline inspection gadget (PIG) body with bypass

Ricardo Ramírez1, Max Dutra2

1 Mechanical Engineer, Department of Mechanical Engineering and Mechatronics, Universidad Nacional de Colombia, Colombia. reramirezh@unal.edu.co

2 Mechanical Engineering Program, COPPE, Federal University of Rio de Janeiro, Brazil. max@mecanica.coppe.ufrj.br


ABSTRACT

A pipeline inspection gadget (PIG) is a device used to run through pipelines for cleaning, measurement and inspection operations. By-pass is the name for one or a set of orifices allowing flow from back to front of a PIG. It is used for speed control or to improve cleaning operations results. By-pass prevents speed excursions in gas pipelines thereby avoiding damage to the PIG or the pipe. Studies and algorithms have been developed to simulate the dynamics of PIGs running inside pipes. Most studies have been for gas pipelines; these have helped to design some PIG models. This work summarises a particular stage during a PIG design project. This PIG will work in liquid pipelines and can stop in the line for inspection and maintenance tasks. Studying fluid force on a PIG when it has low or zero speed is needed for evaluating project feasibility and conceptual design. Selecting a PIG form to minimise the force of interaction with fluid allows a low weight design and lower energy loss.

Keywords: CFD, fluid structure interaction, optimisation, PIG, pipeline.


Introduction

Pipeline inspection gadget/gizmos (PIG) are self-contained devices, tools or vehicles moving through pipelines which are used in cleaning, sizing and inspection (PPSA, 2010) in petroleum and gas pipelines. PIG speed varies due to tube irregularities, changes in physical conditions between contact surfaces and variations in flow. Speed changes are greater in gas pipelines due to fluid compressibility and become dangerous for PIG or tube integrity. High acceleration phenomena are known as speed excursions Mattheus et al, 2007). The use of so-called by-pass orifices in a PIG' s body is one way to reduce speed excursions.

A bypass allows fluid passage from back to front of a PIG and the device' s displacement speed is different from average flow speed. Several studies have been aimed at designing PIGs having a by-pass (Nguyen et al, 2001; Boe, 2007; Korea Gas Corporation, 2010; Podgorbunskikh, 2008). In some cases, by-pass holes are used to form jets of the same fluid to enhance cleaning results; in other cases, a valve is added to the orifice and allows a PIG’s speed to be controlled (Frota, 2007).

A project has been designed for a PIG having special features like the ability to control speed and stop within a pipeline whilst fuel continues to be carried.

The starting concept for a PIG' s body is a cylinder having a central hole to act as a by-pass; the body is supported by polyurethane discs similar to those used in current cleaning PIGs. A valve is placed in the hole for controlling speed, stopping and starting; valve design was not included within the scope of this work.

The aim was to optimise cylinder shape and dimensions to achieve two goals: minimising the interaction force between fluid and body and maximising the body' s internal volume. The first goal would reduce mechanical design and control valve requirements; furthermore, it would reduce energy loss caused by the presence of a PIG in the pipeline. The second goal would create greater available internal volume inside the body so that items such as batteries, electronic boards and actuators could be placed there.

Procedure

Shape optimisation was performed in two stages. The first considered the by-pass hole having a constant diameter and optimum diameter was calculated whilst the second one involved the orifice profile being modified to correspond to various proposed forms for further reduction of the interaction force. The best profile and its optimal size were selected.

It was found during optimisation for the two stages, when varying PIG body dimensions, that if the interaction force was reduced then there was also a decrease in body volume. This meant that if a target were close to its optimum value when far away from its goal, the there were conflicting objectives and optimal diameter value was obtained by using the Pareto frontier concept (Araújo et al, 2009).

Materials and methods

General specifications

The model being studied consisted of a pipe segment and a PIG (Fig. 1). The pipe was cylindrical, straight and horizontal. Nominal project diameter was 12 inches, corresponding to 304.8 mm inner pipe diameter and 323.85 mm outer diameter having 4 m length and 25 micron internal roughness, typical of new steel pipe (Stukenbruck, 2008).

The model used to simulate a PIG was a cylinder having external protrusions simulating the scraping disks and had an axial hole whose shape and dimensions varied for the different cases to be simulated. The outer diameter corresponded to pipe diameter. The PIG length used was twice pipe diameter; this ratio is commonly used by PIG manufacturers and ensures that a PIG does not get stuck in pipeline bends.

The fluid selected was gasoline having r = 747 kg/m3 constant density and m = 6.57x10-4 Pa·s constant dynamic viscosity. The 5 m/s maximum flow speed corresponded to the flow speed typically used in pipelines carrying refined products.

Calculating optimum orifice diameter

Orifice diameter (DO) varied from 10% to 70% of the inner diameter of the pipe (DIT) in 10% steps. The force (F) exerted by the fluid on the PIG body in the direction of displacement was evaluated for each case using a computational fluid dynamics (CFD) tool. The result of force regarding DO/DI T ratio is presented in Figure 2 .

PIG body (VP) volume was also calculated for each case (Eq. 1).

[1]

where, DE was the outer diameter of the PIG and LP its body length. The result is presented in Figure 3.

By increasing diameter, the force decreased and volume increased. As the aim was to minimise force and maximise volume, it was concluded that the two objectives conflicted. Orders of magnitude for force and volume were not comparable. Two functions f1 and f2 were defined for optimal diameter (Equations 2 and 3):

[2]

[3]

Function f1 normalised force to a range from 0 to 1; f1 had to be minimised to minimise force f1. Function f2 normalised volume to a range between 0 and 1 to be comparable to f1 and f2 had to be minimised to maximise volume.

A new function was defined (Equation 4):

[4]

C1 was design criterion by which the designer could give more or less importance to the objective of minimising force or maximising volume. C1 varied from 0 to 1 in 0.1 step. If C1 was small, it meant for the design that it was more important to maximise volume than minimise force and if C1 were large it meant the opposite.

Optimal diameter minimised function f for each value of C1. The results were presented on the Pareto frontier (Figure 4).

A criterion was selected that balanced the two objectives; it was C1 = 0.5. The optimal diameter for minimising the f function was 0.06 m or 60 mm.

Selecting and optimising orifice profile

Using a body having constant by-pass diameter caused the flow to have two abrupt changes in area: sudden contraction at the rear of the PIG and sudden expansion in front of it. Load loss could be reduced by modifying orifice profile to make contraction and expansion gentler.

Results

The force in all cases was calculated using CFD software. Volume was calculated using a geometric formula in the first profile and by integrating the solids of revolution for the second and the third profiles.

Results for the conical profile

The results of force and normalised volume function for the body for the conical profile are presented in Figure 6.

Results for the nozzle-type profile

The results of force and function for normalised volume for a nozzle-type profile are presented in Figure 7.

Results for the polynomial-type profile

The results of force for the polynomial-type profile are presented in Figure 8. The number of possible profiles was greater due to dependence on the six parameters. A genetic algorithm-type process was performed to obtain profiles having lower force. Sets of values r2, s1 and i2 were randomly generated (Figure 5(c)) and each set was used for calculating and building the corresponding profile. Force was calculated for this set of profiles. Profiles corresponding to cases having smaller force were selected and these profiles' parameters were crossed to create a second generation. Software runs were again made to calculate forces until an optimal profile was obtained.

The results of the normalised volume function are presented in Figure 9. The profiles calculated on the results were consecutively named P1 to P18.

Multi-objective optimisation was applied for each profile, similar to that described for obtaining optimal orifice diameter. An optimal variant for each type of profile was selected for each set of variants that minimised the force and maximised the volume. The value for the design criterion taken was: C1 = 0.5.

Table 1 shows the selected optimal variants with their corresponding values for force and available volume.

Analysis of results

Any one of these options provided a solution for building a PIG body. Each one provided the best combination of minimum force and largest available space for its particular type. Three models were built in the prototype construction stage and one was selected from experimental test results.

Also, each one may have been suited to one type of pipeline or fluid. The same criteria regarding minimum force and maximum volume were used to select one of three options for this work; hence, the definite choice for the PIG body was one having an orifice with a nozzle-type profile.

Proposed PIG size as a result of optimisation is presented in Figure 10.


References

Araújo A., Martins, P., Mota Soares, C. M., Mota Soares, C. A., Herskovits, J., Damping optimization of viscoelastic laminated sandwich composite structures., Structural and Multidisciplinary Optimization, Vol. 39, No. 6, December, 2009, pp. 569,579.         [ Links ]

Boe, K., Pipeline PIG., Patent Application Publication. 0214590 A1 United States of America, September of 2007.         [ Links ]

Frota, C., Inspeção com PIG instrumentado (MFL, geométrico e Inercial) no gasoduto Bolivia-Brasil., Neto. Anais da Rio Pipeline 2007 Conference, Instituto Brasileiro de Petróleo, Gás e Biocombustíveis (IBP), Outubro, 2007.         [ Links ]

Korea Gas Corporation. R+D Division., Inteligent PIG. Details of research. En: http://www.kogas.re.kr/eng/index.jsp. últimoacceso: November de 2010.         [ Links ]

Mattheus, L., Kennard, M. y O'Donoghue, A., Velocity control of pigs in gas pipelines., En: Pipeline Pigging and Integrity Technology, Scientific Surveys Ltd., 3401, Lousiana, Suite 255,Houston, TX 77002, USA, cap. 3, 2007, pp. 35-48.         [ Links ]

Nguyen T. T., Yoo. H. R., Rho, Y. W., Kim, S. B., Speed control of pig using by-pass flow in natural gas pipeline., Proceedings of 2001 IEEE International Symposium on Industrial Electronics, IEEE Industrial Electronics Society, Pusan, Korea, 2001.         [ Links ]

Pipeline Pigging Products and Services Association (PPSA)., Pigging terminology, abbreviations and formulae., En:http://www.ppsaonline.com/pigging-terms.php. último acceso: noviembre de 2010.         [ Links ]

Podgorbunskikh, A. M., Devices for Automated Regulation of the Velocity of InTube Pig Flaw Detectors Review., Russian Journal of Nondestructive Testing, Vol. 44, No. 5, 2008, pp. 343 -350.         [ Links ]

Stukenbruck, S., Notas do curso "Escoamento em Dutos"., Pontifícia Universidade Católica do Rio de Janeiro, 2008.         [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License