Abstract Steam injection is a common EOR method for the recovery of heavy crude. This type of oil, or bitumen, is extremely viscous at the reservoirs standard temperature and almost immobile. As a result, steam and other fluids are injected into these reservoirs to enable the transfer of latent heat to the oil-bearing formation and allow them to flow.
The paper introduces a new and efficient algorithm for optimal steam allocation from a series of generators given a changing delivery target at certain time intervals. The solution is addressing the efficiency of the generators as well as the steam distribution to producing wells as a key component in the design of steam-based thermal operations.
Particle Swarm Optimization (PSO), motivated by the social behavior of bird flocks or fish schooling, was first introduced by Kennedy and Eberhart as a population-based optimization technique. In 2004 Sun, Xu and Feng proposed a new version of PSO, Quantum-behaved Particle Swarm Optimization (QPSO), which was inspired by quantum mechanics and trajectory analysis of PSO mechanism. As a quantum system, characterized as an uncertain system when compared to the classical stochastic systems, every particle can appear at any position within a certain probability, thus enabling the swarm to search in the whole feasible region. Additionally, in the QPSO algorithm there are no velocity vectors for particles, thus it has fewer parameters to be adjusted, which makes it easier to implement.
In this paper, a modified QPSO algorithm was employed to solve the steam generators optimization problem described above. First, efficiency curves are established for each one of the different steam generators. For the group of steam generators considered, the objective was to maximize the sum of their efficiencies by adjusting the generated steam from every generator. Additionally, there are operational and design constraints on the generators such as the minimum and maximum amount of steam generated from each generator, the sum of the steam output from all generators, steam quality, etc. Several experiments were generated and are described in the paper. The results show that the modified QPSO was able to produce an optimum realistic solution significant superior to current practices in the field.