Optimization of Polymer Flooding Using Genetic Algorithm

Abstract

One of the methods to achieve optimal conditions in increasing oil recovery through injecting polymers is by optimizing the parameters that influence the success of injecting polymers to provide information and be considered when determining new schemes for implementation in the future. Optimization to obtain the optimum value of the recovery factor and the best value of the parameters PV, injection rate, injection time, injection pressure, adsorption, residual resistance factor (RRF), and inaccessible pore volume (IPV) using a genetic algorithm with three training ratios and testing 70: 30, 80: 20 and 90: 10 using 1000 datasets. The best value obtained for each parameter is at a ratio of 90: 10, which is the best model with the lowest reference error value with RSME of 0.241104 and MAPE value of 19.1964 classified as good prediction with the value of each adsorption parameter 0.1737 (g/l), injection rate 703 (bpd), injection pressure 1816, IPV 0.2524, RRF 4.8319 and finally the optimum value for recovery factor is 53.9557.