One of the main challenges in power plants and particularly in those using solar energy is the increase in efficiency, trying to operate the facility in those operating points in which, the collected energy coming from the sun (which magnitude cannot be controlled), maximizes the output of the installation (net energy, in general). As solar plants are intrinsically time-varying systems, the maximization of their efficiency leads to continuous changes in the operating conditions, thus being difficult to operate them manually and requiring skilled operators. Moreover, the process efficiency will in general depend on the state of the process and the exogenous disturbances (mainly solar radiation evolution) and thus it is difficult even for an expert to predict all the time what is the best choice to maximize efficiency and profit without incurring in operation risks.

Automatic control systems help to cope with this kind of problems. The implementation in a single computer or in a computer network of suitable control algorithms aimed at controlling
the plants while optimising efficiency and profit and minimising risks is one of the main objectives of the work that is currently being performed at CIEMAT-PSA, that is briefly summarised in this paper.