Abstract
Gas Lift consists of gas injection into the tubing to reduce hydrostatic pressure loss. The oil produced from each well is a function of the gas injection rate. Some wells have unstable behavior by injecting low gas lift flow rates, on the other hand, excessive injection rates causes loss production due to its high friction pressure drop. Therefore the gas lift flow rate must be kept in a bounded interval. The objective of gas lift optimization is to allocate a limited amount of gas to a number of wells in order to maximize oil production or project revenue. Evolutionary algorithms were applied to achieve optimal production rates and decide which well must be closed when the compression capacity is severely reduced.
The FPSO considered in this study is equipped with three compressors to produce 16 wells with gas lift and export gas. Eventually, one or two compressors can have a downtime due to maintenance requirements or failures. In this case, at least one well has to be closed. The algorithm was developed to optimize the production when the platform compression capacity was reduced and decide which well should be shut-in. It takes into account not only the total amount of gas available but also the gas lift pressure, which can change along time.
The algorithm could solve a practical gas lift optimization problem and was able to determine automatically which wells had to be closed whenever the total available gas was not enough to operate all wells simultaneously. In case of total available gas be enough that all wells could produce, the results obtained by this genetic algorithm were equivalent to any ordinary optimization algorithm.
The production optimization is an important issue even during abnormal operational conditions. The water cut increases with the aging of wells and the gas lift flow rate demand also increases. Severe compression limitations are undesirable but they can eventually occur.
