Optimal Scheduling of Multiproduct Pipelines Accounting for Flow Rate Dependent Pumping Costs

VANINA G. CAFARO; CARLOS MÉNDEZ; CAFARO, DIEGO C.; JAIME CERDÁ

Las Vegas, NV

Conferencia; International Conference on Gas, Oil and Petroleum Engineering (GOPE-2016); 2016

United Scientific Group

Multiproduct pipelines transport fuels from refineries to distant distribution terminals in batches. The energy needed to move the fluids through the pipeline is mainly associated with elevation gradients and friction head loss. Commonly, friction loss is the major term requiring pumps to keep the flow moving, and it is strongly dependent on the flow rate. Some studies have been focused on reducing the pumping costs, but none of them has thoroughly considered the head loss due to friction when planning the operations of the pipeline. This work introduces an optimization framework based on mathematical models capable of determining the optimal scheduling of single-source pipelines, rigorously tracking power consumption at every pipeline segment through nonlinear equations. The proposed optimization approach is based on Mixed-Integer Linear (MILP) and Nonlinear (MINLP), continuous-time formulations. Continuous models have demonstrated to be the most efficient way to find the optimal sequence, batching and transportation plan for multiproduct pipelines. In a first step, an MILP model provide the set of batch stripping operations to be diverted to the distribution terminals during every pumping run. Next, a rigorous MINLP model specifies the detailed sequence of individual cuts to be performed by the pipeline operator and the flow rate profile at every pipeline segment so as to minimize the operation cost. Real-world case studies are successfully solved using the framework, which proves to be a useful tool for solving large-scale, nonlinear scheduling problems. Important savings are achieved by keeping a more stable flow rate profile over the planning horizon.