(740c) Optimizing Sand Control in Wellhead with Advanced De-Sander Technology

Banchao Shu, Oluwatosin Oyelakin, Priyanka Shahi, Isaac Snyder, Mike Fredrick, Jeevan Dahal, Certified Holdings LLC- Certified Technologies

Abstract

The well testing equipment usually suffers from severe erosion caused by sand production in conventional and unconventional oil and gas wells. The choke manifold, gate valves, and two-phase/three-phase separators are required to repair and clean up frequently, otherwise, it will cause equipment failure and operation problems. Tremendous time have spent on shut-in wells and repair equipment also.

It is critical to remove the sand from oil and gas production stream and ensure continues operation. The wellhead de-sander can remove sand to protect all the downstream equipment, which enables the easier design and operation of solid-handling systems. This novel multiphase cyclone geometry for wellhead was optimized with Computational fluid dynamics (CFD), and Finite Element Analysis (FEA) methods are presented in this paper. An integrated control system was designed with Ladder Logic for automated and continuous operation.

Methodology

The multiphase cyclonic de-sander geometry was designed and optimized by CFD software ANSYS 18.2. The Volume of Fluid (VOF) model, with the kε-RNG method, was used for multiphase flow simulation.

The nonlinear FEA was used to determine the potential failure locations in the pressure vessels. The failure point was identified in the vessel where the total stress, strain, and elongation exceeds the actual limit of the material. The control system design was designed in Ladder Logic.

Results

A 5000-psi rating sand separation system is designed using Boiler and Pressure Vessel Code (BPVC) Section VIII-Rules for Construction of Pressure Vessels Division 2-Alternative Rules. A cyclonic housing and two sand accumulator are be designed based on this code (Figure 1).

Figure 1 Full diagram of the de-sander system

The separator geometry dimensions and separation efficient can be calculated based on stocks’ law, from the following equations. In this model, the multiphase flow spins through a number N of revolutions in the outer vortex. The value of N can be approximated as the sum of revolutions inside the body and inside the cone (Equation 1):

N = (1/H) (Lb + Lc/2) Equation 1

where

N = number of turns inside the device (no units)

H = height of inlet duct (m or ft)

Lb = length (vertical) of cyclone cone (m or ft).
Lc= length of cyclone body (m or ft)

Lapple developed a semi-empirical relationship to calculate a “50% cut diameter” d_pc, which is the diameter of particles collected with 50% efficiency. The expression is (Equation 2)

d_pc = 9µW/(2𝞹*N*V_i*(ρ_p- ρ_a) Equation 2

where

d_pc= diameter of particle collected with 50% efficiency.

W = width of inlet (m or ft)

V_i = multiphase flow inlet velocity (m/s or ft/s) = Q/WH

ρ_p = density of the particle (kg/m³)

ρ_a = air/fluid density (kg/m³)

µ = air viscosity (kg/m.s).

Theodore and DePaola then fitted an algebraic equation to the curve, which makes Lapple’s approach more precise and more convenient for application to computers. The efficiency of collection of any size of particle is given by (Equation 3)

𝜂 = 1/(1+(d_pc/d_pj)^2 Equation 3

where

𝜂_j= collection efficiency of particles in the jth size range (0 < j < 1)

d_pc = diameter of particle collected with 50% efficiency

d_pj = characteristic diameter of the jth particle size range (in µm).

The overall efficiency, called performance, of the cyclone is a weighted average of the collection efficiencies for the various size ranges, namely (Equation 4)

𝜂 = ∑𝜂_j m_j /M Equation 4

where

𝜂 = overall collection efficiency (0 <𝜂< 1)

m_j = mass of particles in the jth size range

M = total mass of particles.

The cyclonic insert geometry was then optimized by ANSYS 18.2 with multiphase flow simulation based on critics such as higher sand separation efficiency and lower pressure drop. The cyclone separation system can treat up to 100 MMSCFD of natural gas and up to 13,000 bbl/d of liquid with 1-2% of sand. The multiphase simulation result suggests more than 99% sand remove efficiency for a particle size larger than 5 µm.

The FEA analysis was further adapted to determine the failure locations in the pressure vessels. Using the FEA analyses to optimize the shape and the wall thickness of the pressure vessel thereby have prevented failure and reduced the cost of over-engineering in the designed.

Figure 2 FEA analysis for the sand Accumulator

The FEA analysis result suggest the potential failure point in the two-inch water inlet hole (Figure 2). We then have added fillet to reduce the stress result from the sharpen are.

The control system would allow for a maximum 75 psi pressure drop in the de-sander system to ensure a longer product lifetime. The advanced automation control system was implemented in the designed thereby elimination possible human interference of the whole assembly operation.

Conclusion

The optimized wellhead cyclonic de-sander with CFD, FEA technology will be able to separate the gas, liquid, and sand with the highest efficiency of 99 percent which has never been approached before. It will provide great protection for all the downstream equipment and reduce the maintenance cost. The automated and continuous process will reduce the shutdown time, reduce human error, and improve the overall production efficiency.