Capturing the effects of oil price uncertainty in carbon integration network design.

Abstract

Carbon integration is a novel concept that targets the recovery and allocation of industrially emitted carbon dioxide, CO2, streams into CO2-using sinks, with the goal of attaining a source-to-sink allocation strategy that meets a desired carbon dioxide emission reduction target, and an ultimate aim of minimizing the cost of the network, while maximizing any revenue attained. Enhanced Oil Recovery, EOR, is considered one of the most attractive CO2 sink options. CO2 streams that are delivered and injected into EOR sites are often classified as great revenue sources for CO2 supplying entities. Since oil pricing heavily affects the revenue generated from sending captured CO2 streams into EOR sites, and since oil prices continuously vary, this paper studies the effect of oil price fluctuations onto the design of carbon integration networks. Hence, oil pricing has been selected as the main uncertainty parameter, and has been fed into a Linearized Multi-Period Carbon Integration model using stochastic data. Since oil prices vary periodically, this model has been formulated over several time periods, in which the oil pricing parameters are allowed to change over time. Subsequently, the proposed model has been optimized using two different approaches: (1) the Binomial Lattice approach, which primarily utilizes average uncertainties as expected values, and (2) the Multi-Scenario approach, which provides the variables’ values as well as the main objective function of the model as a solution basis after accounting for different scenarios. The performance of both methods has been analyzed and compared using random selection of different scenarios which involves simulating each scenario individually. The results obtained demonstrate that each approach has its own advantages and disadvantages. Sometimes, decision makers may find the information extracted from the average values provided by the Binomial Lattice approach to be suitable; other times, a more detailed set of solutions may be desired through the Multi-scena