This paper considers the problem of daily production optimization for a gas lift oil network under the presence of parametric uncertainty. The objective is to find the optimal distribution of injected lift gas to maximize total oil production from an oil network that contains parametric uncertainties subject to some constraints. Typically, the model-based optimization methods in such processes overlook uncertainty and go along with the optimal solution based on the nominal model. Nevertheless, the effect of uncertainty may lead to infeasibility when implemented in real applications. The proposed scenario-based optimization framework in this paper ensures robust feasibility compared to deterministic optimization. Additionally, the superiority of the method has been illustrated in comparison with the other robust optimization counterparts such as Min–Max NMPC in terms of conservativeness and execution time. The simulation results of the nominal, min–max, and scenario-based optimization are compared and discussed.

本文考虑存在参数不确定性的气举油网日产量优化问题。目标是找到注入提升气的最佳分布，以最大限度地提高石油网络的总石油产量，该石油网络包含受某些约束的参数不确定性。通常，此类过程中的基于模型的优化方法会忽略不确定性，并与基于标称模型的最优解相一致。然而，在实际应用中实施时，不确定性的影响可能会导致不可行。与确定性优化相比，本文提出的基于场景的优化框架确保了稳健的可行性。此外，在保守性和执行时间方面，与其他鲁棒优化方法（例如 Min-Max NMPC）相比，该方法的优越性得到了说明。对标称优化、最小-最大优化和基于场景的优化的模拟结果进行了比较和讨论。