Prediction of gas flow rates from gas condensate reservoirs through wellhead chokes using a firefly optimization algorithm

Numerous empirical correlation models for predicting wellhead flow rates have been proposed. Here we apply a recently developed model based upon extensive data from the Ghawar Field (Saudi Arabia) to the Pazanan 1 retrograde gas-condensate field (Aghajari, Iran). A firefly optimization algorithm is applied to select the optimum coefficient values for that model by minimizing the mean square error between measured and predicted gas flow rates from a wellhead-test data set. The input data to calculate gas flow rate includes choke diameter, gas specific gravity, flowing fluid temperature, upstream and downstream pressure. The models prediction accuracy depends upon the coefficient values applied in its formula. The firefly optimization model was tested with various sensitivity cases applying different values to the key control variables γ and N (number of fireflies in the population). Optimum results in terms of minimum mean square error and rapid convergence was achieved with the control variable values γ = 2 and N = 40. The optimum case achieved with low error values and a level of accuracy that is significantly better than the predictions for dataset using the coefficient values applied to the Ghawar field, suggesting that such model coefficients need to be optimized on a field-by-field basis.