Economical optimization of thermochemical storage in concentrated solar power plants via pre-scenarios

Thermal storage is a key point for the development of concentrated solar power technologies. This article aims to develop a methodology for the optimization, from an economic point of view, of a Concentrated Solar Power plant with thermal storage. It addresses two original aspects: (1) it incorporates a thermochemical storage process; (2) it considers the integration of the power plant's production in electricity SPOT markets that present strong price variations over the day, week, and seasons. The strategies of production defining the Storage/Storage-Production/Discharge phases must be optimized with regard to these variable prices. The relevant economic criteria is no more the usual Levelized Cost of Energy, but the Net Present Value, which considers also the revenues of the plant. The required optimization involves two sets of distinct variables which are optimized simultaneously: the physic variables of the thermochemical storage (defining its stored energy and its thermal power) and the operational/strategy variables, that define non-classic storage/production strategies adapted for the price curves. To take into account the time dependent feature of the problem, the notion of pre-scenarios is introduced, which allow to treat the problem, naturally formulated as an optimal control problem, under the classic setting of differentiable optimization. Under this scope, a solution for the optimal design problem of the plant is proposed. First results are presented for a Californian case. This first step highlights the improvement with respect to the classical production strategy (i.e. one storage discharge after sunset).