Long term hydro scheduling with short term load duration and linear transmission constraints

Abstract

Traditional solution methods for the long term hydro scheduling problem often omit short term phenomena such as daily load variations and transmission constraints. Simultaneously accounting for these factors in a single chronological optimization models, may lead to an overwhelming computational burden. Therefore acceptable approximations, that consider basic short term limitations, but still give the important reservoir control results for long term planning, are important. The paper recognizes that in hydro systems, chronology may often be omitted in the short term, when reservoirs are large enough to make short term fluctuations in volume and head negligible. Then reservoirs may be kept at a constant level in the short term to maintain constant head and are removed from the optimization. We remove this chronology by using load duration curves in the optimization. This simplifies the interacting short and long term time scales and, in addition, it is easy to account for transmission constraints, by using linear (DC) power flow. The paper defines an optimization model using linear programming (LP), but with transmission and load constraints, thereby accounting for important short term phenomena in a long term optimization model. The model is applied to a small six bus test system.