Dynamic optimization of heated oil pipeline operation using PSO–DE algorithm

Abstract

Crude oil, with relatively high viscosity, freezing-point and content of wax, is usually transported by heated oil pipelines (HOP) containing many pumping and heating stations. There are many different selections of operation parameters or values of pumps and heating furnaces, but all these different “selections” can satisfy output task and safety requirements. The power and fuel cost of pumping and heating account for 1–3% of the total energy consumption. For the energy saving purpose, it is necessary to optimize the HOP operations. In this paper a dynamic model of HOP was proposed and used to calculate the soil temperature field outside HOP, and then both the spatial and temporal distributions of pressure and temperature of field inside HOP. Using this calculation method and its results, an optimization model is established aiming at minimizing energy cost of running HOP, with outlet temperatures of each heating station, on–off states of each pump, and their head matrix of delivery as the optimization variables. Then a mixed algorithm combining differential evolution algorithm with particle swarm optimization algorithm is used to solve this model. The optimization results are applied to Rizhao–Yizheng digital long distance HOP (375km). Research shows that: verified by real time data acquired by the SCADA, the relative error of the dynamic model’s result is 4.042%, much less than that of the steady-state model’s result (22.67%). The optimized operation scheme can save 17.59% of energy cost for oil transportation task of 2640m3/h. Energy saving effect is remarkable.