MINLP model for the detailed scheduling of refined products pipelines with flow rate dependent pumping costs

Article

Authorship

CAFARO, VANINA GUADALUPE ; CAFARO, DIEGO CARLOS ; MENDEZ, CARLOS ALBERTO ; Cerda, Jaime

Date

2014

Publishing House and Editing Place

Elsevier

Magazine

COMPUTERS AND CHEMICAL ENGINEERING, vol. 72 (pp. 210-221) Elsevier

Summary Information provided by the agent in SIGEVA

Multiproduct pipelines transport fuels from refineries to distant distribution terminals in batches. Theenergy needed to move the fluids through the pipeline is mainly associated with elevation gradients andfriction head loss. Commonly, friction loss is the major term requiring pump stations to keep the flow mov-ing, and it is strongly dependent on the fluid flow rate. Some studies have been carried out for reducingthe pumping costs in multiproduct pipelines, but none of them has been focused o... Multiproduct pipelines transport fuels from refineries to distant distribution terminals in batches. Theenergy needed to move the fluids through the pipeline is mainly associated with elevation gradients andfriction head loss. Commonly, friction loss is the major term requiring pump stations to keep the flow mov-ing, and it is strongly dependent on the fluid flow rate. Some studies have been carried out for reducingthe pumping costs in multiproduct pipelines, but none of them has been focused on thoroughly consider-ing the head loss due to friction along the pipeline. This paper introduces a novel MINLP continuous-timeformulation for the detailed scheduling of single-source pipelines, rigorously tracking power consump-tion at every pipeline segment through nonlinear equations. Real-world case studies are successfullysolved using GAMS?DICOPT algorithm, which proves to be a useful tool for solving large-scale, nonlinearscheduling problems. Important reductions in the operation costs are achieved by keeping a more stableflow rate profile over the planning horizon.

Show more Show less

Key Words

Friction head lossMINLP approachDetailed schedulingMultiproduct pipelineDICOPT solver

Download or request the full text

http://hdl.handle.net/11336/9196