4. Conclusions
The trifurcation employment has particular characteristics for each project, resulting in different geometries, even when all of them perform the same function. When the values of the diameter ratio, pressure or flow are different, the loss coefficients and the flow variations can have major differences in the results. Therefore, a strict validation of the numerical results of the trifurcation of Gurara - ALSTOM® can only be performed with a reduced model of experimental data. However, the approach based on CFD, carefully carried out, based on good mesh analysis, can be an appropriate choice of turbulence models, boundary conditions and other parameters, where the numerical and theoretical analyses should be strongly based on the literature and on own experience of the research group. The results from the numerical analysis of the Gurara-Nigeria project is calculated at the operating point and beyond in order to have a wide range of results for one later comparison with other numerical and experimental results of pressure losses. However, many results (experimental and numerical) are only calculated at the design point. On the other hand, in order to generalize the results for the particular conditions of each trifurcation, the loss coefficient correlates with the Reynolds number for the steady state, and for the non-stationary state, comparisons were made according to the variation in the total loss coefficient at the time, by the flow. There is very little information related to the loss coefficients in trifurcations in transient flow. In specialized literature two studies are found that report the experimental and numerical results about the behavior of pressure loss: Ruprechs et al [10] and Tate et al [9]. For the results obtained by Ruprechs considering one spherical trifurcation, the experimental data were obtained based on a smaller scale model with an acquisition time of 210 s, as seen in Figure 15.
