Published on January 24, 2008
ALGORITHM RELATES PROCESS PERFORMANCE MATRIX WITH SEPARATORS CONTROL DYNAMICS IN A CRUDE OIL FLOWSTATION MODELLING: ALGORITHM RELATES PROCESS PERFORMANCE MATRIX WITH SEPARATORS CONTROL DYNAMICS IN A CRUDE OIL FLOWSTATION MODELLING E. O. Okeke NNPC R&D Division, Port Harcourt, Nigeria APACT 04, Bath, 26 – 28 April, 2004 INTRODUCTION: INTRODUCTION The Nigerian National Petroleum Corporation (NNPC) is involved in oil & gas upstream and downstream business, The Corporation has joint venture (JV) partners for upstream operations, As a consequence of its drive to increase production in upstream sector, a JV partner required a re-evaluation of a 30Mbpd swamp standard design flowstation. PROJECTED PLAN: PROJECTED PLAN The JV partner then planned a stage-wise programme viz, Production forecast, Flowstation re-evaluation and proposal for modification, Engineering & construction. The JV company produced and supplied the forecast for flowstation re-evaluation, Flowstation re-evaluation formed the foundation of this work, Engineering & construction handled elsewhere PRODUCTION FORECAST - Scenarios: PRODUCTION FORECAST - Scenarios Production forecast for 2002 – 2006: Weak aquifer, Moderate aquifer, Strong aquifer The moderate aquifer the most likely. The weak and strong aquifer each have a probability of less than 10%. FLOWSTATION RE-EVALUATION : FLOWSTATION RE-EVALUATION Stage-wise approach, viz, Data evaluation and review of Basis for Design (BFD). Feed characterization & flowstation configuration, Modeling concept and algorithm Comparison of HYSYS result with original design requirements. Sensitivity analysis to determine the highest permissible production to be handled. 5 sets of separator liquid levels were considered: 50%, 55%, 60%, 65%, and 70% Evolution of the performance matrix FEED & FLOWSTATION CONFIGURATION: FEED & FLOWSTATION CONFIGURATION The facility comprises of a single oil and gas separation train, with 3 stages of separation: HP, LP and surge vessel. Well fluids are transported through flowlines which are tied-in to the inlet manifold and enter the process via the production headers. The wet crude is pumped into the delivery line and stabilized at the flowstation to surge vessel conditions, with vessel pressure not exceeding 0.35 bar(g) to minimize degassing at the receiving terminal. Liquid carry-over to flare must be minimized to less than 10 ppm of oil for environmental and conservation reasons. CONFIGURATION OF SEPARATORS: CONFIGURATION OF SEPARATORS SEPARATORS LEVEL CONTROL CONECPT: SEPARATORS LEVEL CONTROL CONECPT THE FACILITY: THE FACILITY PROCESS CONDITIONS: PROCESS CONDITIONS The HP separator operate at 12.0-13.0 bar(g) and the LPs at 2.5-3.0 bar(g). The production split ratio between HP:LP is 95:5. The separators are fitted with vanepack internals and a Schoepentoeter inlet device and can handle 50 Mbpd gross production. There are four 12 Mbpd pumps for crude export. The size of level control valves on both the HP and the LP separators can accommodate 45 Mbpd each. The two crude export PD meters have a capacity of 34 Mbpd, and are efficiently been operated at less than 70% of the rated capacity. CALCULATION FACILITY FLOWSHEET: CALCULATION FACILITY FLOWSHEET TEST Sep 1 LP Sep 3 HP Sep 2 Surge Drum Sep 4 Crude Inlet Manifold QG1max ρL1 λ1max QG2max ρL2 λ2max QG3max ρL3 λ3max QG1, ρL1 QG3, ρL3 QG2, ρL2 QG4, ρL4 QG4max ρL4 λ4max Crude delivery Gas to flare MODELLING–Volumetric Flows: MODELLING–Volumetric Flows The separators are in series, hence for nth separator the generalized standard separator flow equation for highest volumetric gas load factor, Q*nmax is Q*nmax = QGnmax Gn /(Ln - Gn ) (1) QGnmax is the highest envisaged gas flow rate which includes a margin for surging, uncertainties, etc Ln and Gn the densities of liquid and gas. MODELLING –Maximum Gas Load Factor: MODELLING –Maximum Gas Load Factor The minimum required vessel cross-sectional area for gas flow, is: AGnmin = Q*nmax /nmax (2) From equations 1& 2 λnmax = (QGnmax Gn /(Ln - Gn ))/ AGnmin (3) nmax is the maximum allowable gas load factor, which is a measure of the gas handling capacity of the selected separator. Hence the maximum allowable gas load factor for the flowstation, which determines the maximum crude that can be handled, becomes Єmax = maximum (λ1max,, λ2max,…, λnmax) (4) BOUNDARY CONDITIONS: BOUNDARY CONDITIONS No liquid carry over to the main flare system and surge vessel pressure not exceeding 0.35 bar, Separator temperature and pressure at 37.8 C and 13 bar respectively, For vertical vessels the wire mesh demisters efficiency is assumed to increase at gas flows less than 30% of design throughput, if the droplet size distribution of the liquid entrained in the gas flow remains the same. For horizontal vessels wire mesh demisters efficiency of the vane pack mist eliminators are assumed to decline at a gas flow of around 30 - 50% of the design throughput. The crude oil pump transfer capacity of 67 m3/hr each. The maximum value for the gas load factor of 0.15. THE APPLICATION OF HYSYS: THE APPLICATION OF HYSYS All equipment, process lines and pipeline items modeled and installed for simulation as per the flowstation design and operations requirements. HYSYS.Plant 3.1 applied for entire flowstation steady state simulation HYSYS FLOWSHEETING: HYSYS FLOWSHEETING HYSYS RESULT VERSUS AS-BUILT DATA: HYSYS RESULT VERSUS AS-BUILT DATA Facility material balance HYSYS VS AS-BUILT ANALYSIS: HYSYS VS AS-BUILT ANALYSIS Testing of the configuration on HYSYS with the original design requirements was satisfactory. The gas load factor obtained with HYSYS calculation was within the value defined for the horizontal vessel of this type fitted with Schopentoeter inlet device, showing that the concept and the algorithm are applicable to any other rigorous analysis. PERFORMANCE EVALUATION PARAMETERS: PERFORMANCE EVALUATION PARAMETERS Parameters are yj, gas load factor, xi, crude oil density zj, HP separator liquid level, and aij, crude oil flowrate: i=1,….,n; j=1,….,m. Generalized performance matrix for zj, crude oil density, HYSYS ADJUST OPERATION: HYSYS ADJUST OPERATION The ADJUST operation in HYSYS was used to determine maximum crude oil handling capacity for selected gas load factors using equations (1) to (4), viz, For a given crude oil density, zj For HP separator liquid level xi, Specify gas load factor, yi Carryout HYSYS simulation, solve equations (1-4), determine maximum crude handling capacity, aij, Vary xi (xi=50%,55%,60%,65%,70%), repeat from (c), Vary yi (yi=0.10,0.12,0.13,0.14,0.15), continue from (b) Vary zj, (655,837,856, 871,907) and continue from (a) Secant/Broyden method employed for solution. PERFORMANCE MATRIX : PERFORMANCE MATRIX A performance matrix evolved for range of crude oil flowrate for the simulation. The performance matrix shows the relationship between the gas load factor, the liquid level in the separator (here HP separator) and the flowstation maximum crude handling capacity for a given crude oil density. PERFORMANCE TABLES: PERFORMANCE TABLES PERFORMANCE PROFILES: PERFORMANCE PROFILES PERFORMANCE ANALYSIS: PERFORMANCE ANALYSIS For given crude characteristics, The trend in the performance matrices showed consistently that the crude oil handling capacity increases with the decrease in liquid level in the separator. For same liquid level, crude oil handling capacity increases with increase in gas load factors. Separator configuration and geometry determined the maximum gas load factor permissible and hence the crude oil handling capacity of the stations. APPLICATION OF ALGORITHM: APPLICATION OF ALGORITHM Since liquid level controllers are installed, the performance matrices can help an operator determine the set points for the separators in a flowstation of this type and set the controller to maintain this level. For crude oil of given API, the controller can be set to maximize crude oil handling capacity, taken into consideration, the flowstation’s process limitations, environmental and downstream operations requirements. CONCLUSION - 1: CONCLUSION - 1 The study has shown that the flowstation of this nature can be modeled with all the process requirements determined. With the impact and implication of the gas load factor on flowstation performance and capacity utilization identified and incorporated into this algorithm, it is now possible to optimize the capacity utilization and performance of this flowstation. CONCLUSION - 2: CONCLUSION - 2 Performance matrices show consistency in the response of the flowstation to changes in crude properties and separator liquid level. Since the flowstation was constructed based on a standard design, this algorithm can automatically be applied to other flowstations of same standard design and possibly handling of crudes of different properties. This strategy will enable the company (the operator) to carry out quick flowstation performance evaluation for differing crude oil characteristics for specified gas load factor.