Finite horizon electricity pricing for investment recovery in renewable-based power system

The increasing penetration of renewable electricity generation is transforming the electricity sector adding complexity to the bidding and electricity prices estimation processes. It also shifts overall cost sensitivity from operation (fuel) costs to investment costs (as in the telecommunication sector). In this context, cost minimization models for capacity expansion are very often based on the property that, for a perfectly adapted system including non-served energy, remuneration based on the marginal system cost allows companies to recover their overall operation and investments costs. These models are usually formulated as finite-horizon problems when they should be theoretically solved for infinite horizons under the assumption of companies infinite lifespan. However, infinite horizon optimizations cannot be approached with mathematical programming that requires a finite set of variables and constraints. To overcome this drawback, several approximations have been proposed in the literature based on finite horizon models that tend asymptotically to the original infinite ones. In some of these approximations, the investment costs are annualized along the plants lifespan and include a residual value that represent these costs along the periods beyond the truncated infinite
horizon. This paper proposes a novel approach to annualize the investment costs for finite horizon models that needs neither residual values nor non-served energy, but that provides the electricity marginal costs that exactly coincide, under some reasonable hypotheses, with those of the original infinite horizon model. The proposed model can be used for longterm electricity pricing in fully renewable systems when non-served demand is not allowed, as occur in many current power systems, guaranteeing investments costs recovery.