The Influence of Friction Coefficient of Particles on Numerical Simulations

ABSTRACT

The taking place of landslides in seabed is a happening occurring on several oceanic surroundings. These phenomena have triggered damage and destruction of just offshore constructions. To try understanding somewhat more these phenomena today's work used a free of charge tool where landslides could be simulated trough DEM coupled to computational fluid dynamics. The used software in simulations was CFDEMcoupling. Through this software and just offshore landslides were simulated onshore, supposing a 6? slope. To each simulation performed an alternative particle coefficient of friction was used and the modifications on average level and outflow were evaluated. The info show a pattern of development in maximum average elevation of outflow in parts near of the start of activity and the loss of this elevation in regions further right from the start of outflow as the coefficient of friction of contaminants increases. These conducts take place in onshore so far as offshore flows, the utmost average levels with the same coefficient of friction in debris are close if the ideals between onshore and just offshore flows are likened.

FOREWORD

Because the first results of oilfields on continental slopes several studies have been produced by experts worldwide to try understanding the geotechnical-geological conditions and the energetic of just offshore sediments in these areas. Slopes instabilities must be cured as a potential risk, upon this bank account these may appear on areas where exploitation and development of coal and oil equipment is installed, for instance, just offshore oilfields, anchoring system from floating constructions, move and pipelines pipes [1].

The just offshore landslides are consequence of a complex relationship of several factors that respond together at different intensities and scales of their time, corresponding with the geology of the website where the field is situated. The geological-tectonic-oceanographic environment regulates the action of every balance factor of sea ground and creates, or not, a potential instability on site [2].

Corresponding Locat and Lee [3], just offshore landslides are started out by the increasing on the performing tensions, lowering on material amount of resistance or for both. The major lead to systems of natural process are associated to surprise waves made by earthquakes and cyclic surcharge created by the waves and storms.

The just offshore landslides show different degrees of density, elope fluid denseness and surrounding smooth thickness. In these moves, called thickness flux, the allergens will move by the action of gravity and a activity is manufactured by this movements of interstitial liquid [4]. Boffo [5] says that density flows are the key way of transport and deposition of material in aqueous medium.

The in situ view of just offshore landslides is very hard and also simulates these slides on physical models; the numerical simulation becomes an important tool to permit an improved knowledge of the behavior of the thickness fluxes. The happening is dynamic which involves materials with different densities the modeling with discrete elements combined to liquid dynamics. In this intensive research the free software CFDEMcoupling was used, the program is produced by DCS Processing GMBH [6, 7, 8], which allowed executing simulations using discrete elements to simulate the particle stream and the smooth dynamics to simulate the encompassing fluid.

MATERIALS AND METHODS

A complete of ten numerical simulations were carried out, five onshore and five just offshore, each simulation with some other coefficient of friction (0.54; 0.60; 0.70; 0.81 and 0.85), all the moves of sole pulse.

COMPUTATIONAL TOOLS

The simulations were performed with CFDEMcoupling software, which synthesize the computational substance dynamics (CFD) and the discrete aspect method (DEM) to model fluid-granular combined systems. The movements of debris is received using DEM, which can be applied Newton's regulations of motion to all or any particles, as the CFD is employed to compute the fluid move through Navier Stokes' equations. The relationships between the liquid and solid stages are modelled through the Newton's third legislations.

The Hertz-Mindlin simplified model was used in the DEM research and an initial order model with an incompressible substance was used to CFD.

The CFDEMcoupling software was installed in a Desktop pc with a Intel(R) CoreTM i7 -4770 (8 MB Cache, 3.4 GHz) processor chip, 16 GB Memory (DDR3, 1600 MHz), Hard disk drive 1 TB (SATA 7200 rpm) jogging operating-system Ubuntu 12.04 LTS (Precise Pangolin) 64 bits.

TEST BOX

To be able to simulate the moves of gravity a test package, with measurements 0.70 m x 0.20 m x 0.50 m (span x width x elevation), was idealized. With this test box there's a slope of 6? and a bay (0.10 m x 0.20 m x 0.30 m) m? in volume to reserve the particles. The connection between your slope and the bay is manufactured by way of a floodgate of area 0.20 x 0.10 m?. Underneath of the slope works as a trench, therefore the contaminants can stay there after their passing through the slope. Shape 1 represents at length the plan of the test package.

PARTICLES

Through the simulations spherical allergens of diameter 0.003 m were used. Selecting particle diameter was done seeking limit the utmost number of allergens inside the bay (particle tank) of test package in about 310,000 (3 hundred and ten thousand). The original packing of debris inside the bay adopts a cubic composition of focused face (CFC), which provided a thick arrangement of debris.

To execute the just offshore and onshore simulations, it was essential to define the next particle properties: Young's modulus, Poisson's percentage, restitution coefficient, friction coefficient. In the purpose of observe the modifications happening in the simulated moves of materials with different coefficient of friction, simulations using assorted values of the property were completed, both in onshore as just offshore conditions, the rest of the properties values remained fixed. In Stand 1 the followed prices to each particle property, in simulations, is shown:

SIMULATION DEVELOPMENT

Following the creation of particle deal inside the bay, the floodgate was removed and the contaminants were released to perform off within the slope. In simulation the full total time was of 5 secs onshore, the period between iterations was of 0.00001 second. The info was record in intervals of 0.01 second. In just offshore simulations, the full total time of simulation mixed because instabilities of the operating-system Ubuntu, nevertheless the times come to allowed identify the conducts appealing of today's research.

DETERMINATION OF MAXIMUM AVERAGE HEIGHT OF FLOW

The slope inside the test field was evenly divided in 5 parts to assist in the perseverance of average elevation of move, as shown in Body 2.

The willpower of average elevation stream on those parts the levels of flow at the start (hi) and the finish (hf) of every region were examined as shown in Number 3, the common height of movement is the arithmetic mean of the moves at the start and the finish of every section. The common heights of flow were established in each region as soon as of type of the first particle in the analyzed region up to the ultimate time of simulation.

RESULTS AND DISCUSSION

Analyzing the utmost average height come to by the moves in each section through the movements, was possible to sketch the onshore circulation contours (Physique 4) and from just offshore flows (Amount 5). Notice, in Amount 4, that in your community near to the start of circulation (Region 1) the utmost average elevation of move grow whit the increasing of the particle friction coefficient. Inasmuch, in your community further right from the start of stream (Region 3), the utmost average elevation of flow grows up with the decreasing of particle friction coefficient. Along others areas occurs a change of the move behavior with regards to the utmost average level. This action is repeated in just offshore flows, Number 5, however the difference between the flows in Regions 1 smaller than the one in other onshore flows clearly.

Once the onshore and just offshore flows are likened (Physique 6), both maximum average level are incredibly close. The just offshore flows on areas further the start of motion show a pattern to get average levels smaller (Parts 5 and 3). Inside the other hand, in your community closer of the start of movement (Region 1), only in circulation with the friction coefficients of ideals 0.70, 0.80 and 0.85 the last behavior was detected. In moves with friction coefficients of 0.54 and 0.60, the just offshore flows come to maximum average levels higher than the onshore moves with the same worth of coefficient.

Note that there is certainly, in both just offshore and onshore moves, a propensity of the contour of maximum average elevation have a tendencies approximately linear. Out of this observation, the persistence coefficient (R?) was done using linear regression of every data flow, for each and every flow examined, as shown in Desk 3.

Founded on the full total results, was possible determine equations that symbolizes the deviation of the utmost average height over the movement, considering different friction coefficients in allergens to both onshore (formula 1) and just offshore (formula 2) flows.

These equations shown good approximation of the bought data by simulations, they permit someone to forecast of the utmost average elevation of flow across the slope to any friction coefficient in debris. It really is clear these equations are constrained by other variables found in simulations as: Young's Modulus, Poisson's proportion, restitution Coefficient, inclination of slope, etc.

CONCLUSION

Predicated on the performed simulations was possible verify an increase in the utmost average level occurs in parts near of the start of the circulation as the friction coefficient of allergens is incremented. Alternatively, there's a tendency of lower on the utmost average level of the circulation as the friction coefficient boosts in regions definately not the start of the flow. Those actions are found both in onshore and just offshore moves.

Also there's a great similarity among just offshore and onshore curves of moves, examining both at the same friction coefficients, showing that the encompassing fluid, normal water, creates a tiny disturbance on maximum average level of the movement. Also was possible to see a linearity propensity in all curves to the utmost average level, which empowered define equations that depicture the habits of maximum average level contours in line with the friction coefficient of contaminants.

ACKNOWLEDGMENT

The authors wish to say thanks to the Universidade Estadual do Norte Fluminense Darcy Ribeiro (UENF), the Funda??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES) and Petr?leo Brasileiro S.A (Petrobras) for financial support in the study.

REFERENCES

Hance, J.J. (2003) Development of a database and assessment of seafloor slope stability based on published literature. Dissertação (Mestrado em Engenharia) – Austin – TX, The University of Texas at Austin – OTRC/UT, 245p.
Silva, B.A. (2005) Análise de estabilidade de taludes ao longo da rota de um duto submarino na Bacia de Campos, RJ. Dissertação (Mestrado em Engenharia Civil) – Rio de Janeiro – RJ, Pontifícia Universidade Católica do Rio de Janeiro – PUC-Rio, 117p.
Locat, J., Lee, H.J. (2000) Submarine landslides: advances and challenges. Proceedings of the 8th International Symposium on Landslides. Cardiff: A.A. Balkema, v. 1: 1-30.
Middleton, G.V., Hampton, M.A. (1976) Subaqueous sediment transport and deposition by sediment gravity flows. In: Marine Sediment Transport and Environmental Management. New York, Willey: 197-218.
Boffo, C.H. (2010) Caracterização física de correntes de densidade em ambientes salinos sob diferentes contrastes de densidades. Dissertação (Mestrado em Engenharia) – Porto Alegre – RS, Universidade Federal do Rio Grande do Sul – IPH/UFRGS, 121p.
Goniva, C., Kloss, C., Pirker, S. (2009) Towards fast parallel CFD-DEM: An open-source perspective. In: Proceedings of Open Source CFD International Conference. Barcelona, Spain.
Goniva, C. Kloss, C., Deen, N.G., Kuipers, J.A.M., Pirker, S. (2012) Influence of rolling friction on single spout fluidized bed simulation. Particuology, DOI 10, 1016 / j.partic.2012.05.002.
Kloss, C., Goniva, C., Hager, A., Amberger, S., Pirker, S. (2012) Models, algorithms and validation for OpenSource DEM and CFD-DEM. Progress in Computational Fluid Dynamics, 12, Nº 2/3, p. 140-152.