Investigation into Effect of Burial Depth of Offshore Pipelines in Tandem Position on Flow Separation Using the Experimental and Numerical Models

1. he ever-increasing need to fossilized fuel has caused a rapid development in offshore industries. One of the most important offshore structures that have extensively been used for transporting oil and gas products is pipeline. On the other hand, fatigue of these structures mainly causes huge evil economic and environmental consequences. The pipelines lying at sea bed normally cause variations in flow pattern around them and also create eddies in the front- and backward of pipeline which is along with an increasing in turbulence intensity 1]. These variations can lead to the fatigue in structure or scour around it and eventually threaten structural stability [2]. In this paper, in order to consider the action of pipe on flow, the caused variations on flow pattern around the semi-buried offshore pipelines due to steady current have experimentally and numerically been investigated.2. 2.1. Experimental studyGenerally, the main aim of doing the laboratory tests was to find out the separation lengths at the upstream and downstream of the pipelines in tandem positions for different burial depth-diameter ratios (i.e., G/D). The experiments have been conducted in the Hydraulic Research Laboratory. For the experimental section, a number of tests have been carried out in a flume with 10 meters length, 0.3 meters width and 0.5 meters depth using P.V.C pipes with 6.35 centimeters in diameter at steady state flow condition and for different burial depth-diameter ratios (i.e., G/D) and for the single and double pipelines. To visualize the flow patterns, the polystyrene particles with 1.05 gr/cm3 in density have been used (Fig. 1). In order to help to physical understanding of the phenomenon, the whole processes of tests have been recorded using a digital camera.2.2. Numerical modeling For the numerical modeling, the flow field has been analyzed using a computational fluid dynamics software, called FLUENT. For this software, the governing equations have been discretized using the finite volume method. The algebraic equations have then been solved using simple pressure-velocity coupling method. For the turbulence model, the two-equations k-ε RNG model has been deployed to calculate the Reynolds stresses. Geometry and mesh generation processes have also been done using a preprocessor software, called GAMBIT. The used discritized methods and under relaxation coefficients are shown in Tables 1 and 2, respectively.3. The numerical model results were first verified against the experimental and numerical model results of the other researches and it was found that the simulation model results of the current study were in good agreements with the experimental data of the other researchers. Then, the results of numerical models were compared with the test data of the current study and some relatively good agreements were achieved (Table 3). In this regard, it was found that the maximum discrepancy between the experimental and numerical model results for the reattachment length was generally less than 17% and only in two cases for single pipeline it increased up to 25%.4. The main results of this research study may be summarized as: 1. Three separation zones (one at upstream and two at downstream) around a single pipeline were generally observed.