(84b) A New MILP Model of Multi-Product Pipeline Distribution By Handling Transmix with a Combined Method of Blending, Distillation and Downgrading

Pipeline is much more efficient and reliable to transfer large amounts of petroleum products than other ways via tanker truck, rail cars, and vessel. To save capital and operating costs, one pipeline needs to transfer different types of oil products to different oil depots according to their ordinary demands. As two consecutive different types of oil products inside a pipeline will not be separated by a physical divider, the transmix (TM) will be generated in the interface between oil products. To guarantee the oil quality of product adopted by storage tanks at depots, a considerable amount of TM product from the pipeline needs to be handled appropriately at depots. In reality, TM can be handled by blending, distillation and downgrading. The costs for blending and distillation comes from the storage, blending operation, transportation and reprocessing, while the cost for downgrading is a loss of margin which comes from the different price between the two mixed refined products. Though the downgrade operation and cost is easier and less than blending and distillation, only the TM which consists the same type of products with different grades (i.e., premium gasoline and regular gasoline) could be handled by downgrading the upper grade product to the lower grade one (i.e., downgrade premium gasoline to regular gasoline). Conceivably, the multiproduct pipeline scheduling problem with consideration of handling with TM inside the pipeline present a big challenge to oil industries. There are few works reported in this area (e.g., Rejowski et al. (2003); Cafaro and Cerdá (2004 and 2010)). However, there are still lack of studies on dealing with the TM .

In this work, a continuous-time based MILP model is developed to optimize the transportation and distribution of multi-product, which includes a long-distance pipeline and several depots which consists of storage tanks and transmix tanks dedicated to hold pure product and TM, respectively. The TM generated inside the pipeline will be handled with a combined method of blending, distillation and downgrading. The TM with different types (i.e. gasoline and diesel) will be handled with a hybrid method by blending product-rich mixture into the pure product and distillation of the well-mixed part of the TM; while the TM with the same type but different grades oil products (i.e., premium gasoline and regular gasoline) will be handled by downgrading the upper grade product to the lower grade one. Overall, the scheduling model could provide optimal solutions including injection sequence and volume of product into pipeline, product movements inside pipeline, discharging product to multiple depot tanks, as well as sending out oil product based on consumer demands in the continuous time formulation base. The efficacy of the developed scheduling model is demonstrated by multiple case studies.

Keywords: MILP, Pipeline, Multi-product, Scheduling, Transmix, Blending, Distillation, Downgrading

References:

Cafaro, D., & Cerdá, J. (2004). Optimal scheduling of multiproduct pipeline systems using a non-discrete MILP formulation. Computers and Chemical Engineer- ing, 28, 2053–2068. ISSN 0098-1354.

Cafaro, D., & Cerdá, J. (2010). Operational scheduling of refined products pipeline networks with simultaneous batch injections. Computers and Chemical Engineer- ing, 34, 1687–1704 .

Rejowski, R., & Pinto, J. (2003). Scheduling of a multiproduct pipeline system. Com- puters and Chemical Engineering, 27, 1229–1268 .