## Field development from subsurface to topside optimization - Case Study: Songo Songo Gas Field, Tanzania

##### Master thesis

##### Permanent lenke

http://hdl.handle.net/11250/2615136##### Utgivelsesdato

2018##### Metadata

Vis full innførsel##### Samlinger

##### Sammendrag

This thesis is focused on identifying, analyzing, and evaluating the possible development concepts for optimal development of the Songo Songo Gas field, Tanzania through reservoir and surface network integration. The field is currently producing therefore the main objective of this work is to suggest alternative development strategy which could be employed for the field to produce at a rate of 200MMScfd and maximizing the project's expected net present value (NPV).The estimation of reserves is significant for field development. In this study, volumetric analysis method has been used to estimate the gas reserve by applying Monte Carlo simulation to compute Gas Initially in Place (GIIP), and Total Recoverable Reserve (TRR). Also, Monte Carlo is used to perform economic analysis that decide the best production strategy in which P10, P50 and P90 NPV s are obtained. The field optimization is done by integrating reservoir and surface network model in which the simulation is performed by setting the field gas target flow rates 200MMScfd.The reservoir analysed are Neocomian and Cenomanian formation which found in Songo Songo gas field. Six development cases (options) are conducted in this study, these are Case 1 with 8 wells (6 Neocomian and 2 Cenomanian), Case 2 with 6 wells (4 Neocomian and 2 Cenomanian), Case 3 with 5 wells (4 Neocomian and 1 Cenomanian), Case 4 with 9 wells (7 Neocomian and 2 Cenomanian), Case 5 with 9 wells (6 Neocomian and 3 Cenomanian), Case 6 with 8 wells (5 Neocomian and 3 Cenomanian). The 6 and 12 inches flowlines were used for every case which make total of 12 cases.The total field GIIP obtained for P10, P50 and P90 is 2141.69 Bcf, 1779.50, and 1479.16 Bcf respectively. The total Recoverable Reserves obtained for P10, P50 and P90 is 1409.15 Bcf, 1157.68 Bcf, and 959.45 Bcf respectively.The best option for developing P90 (1P) reserve is case 4 based on P10 NPV and case 1 is best option based on P50 and P90 NPV s. For P50 (2P) reserve development, the best strategy is case 4 for P10 NPV and based on P90 NPV and P50 NPV the best strategy is case 1, the best option for developing P10 (3P) reserve is case 4 for P10, P50, and P90 NPV. The sensitivity analysis conducted on number of wells and separator pressure in which 4 and 5 cases were used respectively to determine their effect on plateau length. For number of wells Case I, II, III, and IV with 2 (1 well per formation), 3 (1 Neocomian and 2 Cenomanian), 3 (2 Neocomian and 1 Cenomanian), and 4 (2 Neocomian and 2 Cenomanian) wells respectively. Also, for separator pressure cases developed are case A, B, C, D, and E for 10%, 20%, 30%, 40% and 50% of the separator pressure respectively.The decrease of separator pressure to 800 psig (55.16 bar) equivalent to 50% of the current value of 1600psig (110.32bar) is the best approach to extend plateau length for all options of developing P10, P50 and P90 reserves, the value yielded high gas recovery and maximum NPV for all category of reserves (1P, 2P, 3P). Therefore, Case E is the best option for extending of plateau.There is possibility of hydrate formation to avoid this problem in the pipeline the Pressure and temperature of the fluid flowing in the pipeline was kept outside the hydrate formation region. There were no corrosion and erosion problems in the pipe.The study concluded that the development of P10, P50 and P90 reserve categories is selected according to the NPV percentile (P10, P50, P90) whereby case 1 and 4 are the best options for P50 (2P) and P90 (1P) reserve, for P10 (3P) is case 1. The 12 inches flowlines should be used, and it is revealed that the separator pressure have high impact on plateau length compared to other parameters. Finally, it is recommended that the number of iterations (Number of Samples) in Monte Carlo simulation suggested to be more than 2000 to decreases the error percentage as well as increasing of the perforation intervals in the gas producing layers to maximize reservoir exposure.