The giant ONGC Mumbai-High offshore field produces about 250,000 bopd (60% water-cut) from 700+ wells through 110+ unmanned production platforms. The production platforms are connected by subsea flow lines to 6 processing platforms where the oil, water and gas are separated. More than 80% of the wells in the field are on gas-lift.

This paper discusses an integrated-network-modeling-and-optimization approach to overcome known drawbacks of single well gas-lift optimization. The new higher fidelity model captured the interactions between wells, flow-lines and platforms, field-wide. Using this approach, previously hidden opportunities for significant production improvement were identified. The model indicated a possibility of significant production increase (2 to 4%) and production efficiency improvement (significant reduction in lift gas requirement), which could be achieved by better allocation of lift gas and repair and redesign of the sub-optimal gas-lift wells. The latest available field implementation results are also presented.

Currently, in the absence of full-loop control-automation of the gas-injection rates/pressures, logistical challenges made it impractical to simultaneously implement the calculated optimum settings field-wide. The network interactions are highest at the production-platform level between adjacent wells and connecting shared flow-lines. So, from practical implementation considerations, optimum gas-injection rates were calculated and simultaneously implemented at the production-platform level, one (or a group of few connected) production platform at a time, with little or no compromise on the quality of the solution or on the validity of the approach.

To date, actual optimization implementation changes in 14 of the 110+ platforms (160 of the 700+ wells) indicated immediate production increase of over 2000 bopd and significant reduction in lift gas requirement. Based on these early indications, it is envisaged that the field-wide oil production gain can possibly be as high as 6%.

Additionally, this paper discusses the challenges and recommended approaches with regards to sustainable model maintenance, solution implementation and workflow. Status of control automation and on-line modeling plans (planned for implementation in future) are also discussed.