Comments on: A comparative study of time aggregation techniques in relation to power capacity-expansion modeling 2
-
Ken McKinnon [1]
-
[1] University of Edinburgh
University of Edinburgh
Reino Unido
-
- Localización: Top, ISSN-e 1863-8279, ISSN 1134-5764, Vol. 27, Nº. 3, 2019, págs. 410-413
- Idioma: inglés
- DOI: 10.1007/s11750-019-00522-4
- Enlaces
-
Resumen
Real-world energy planning problems have to deal with uncertain future estimates of costs and technologies, and assess the operational costs of different possible investment decisions with operational model of appropriate detail. The introduction of renewable generation, which is highly volatile, stochastic and dependent on location, drives the need to use operational model of finer spatial and time resolution, and to test them on a much wider variety of operational situations. It is, however, impractical to solve these models in full detail, so it is necessary to select approximations that achieve an appropriate balance between model detail and the computational speed. What is appropriate depends on what the investment problem is. If the plan is for the near future with a small number of possible investment options, then the correct balance may be to use an operational model of high accuracy and high solution time. At the other extreme, multistage planning problems that take into account future uncertainties require the solution of large numbers of operational problems, so the correct balance will be to use faster operational models and sacrifice some accuracy. However, given the inevitable uncertainties in long-term planning, this loss of accuracy in the operational model may not add significantly to the overall uncertainty in the results of the planning model.
