Solar external receivers with molten salt as heat transfer fluid are the most critical subsystem of a Solar Power Tower (SPT). Receiver tubes work under extreme conditions due to the high incident solar flux and the potentially corrosive environments. These demanding conditions of operation usually produce the failure of the receiver by stress corrosion cracking. The unsteady solar flux and the large size of the heliostat field and the receiver make very complicated accurate measurement of the spatial heat flux on the receiver tubes. Hence, modelling accurately the solar flux onto the receiver and the heat transfer in the tubes is required. This PhD thesis consists in the development and validation of several thermal models of external receivers to improve the estimation of the temperature distribution on the receiver tubes and the thermal efficiency. The application of the models has enabled to establish the guidelines for the accurate and safety design of the external receivers. In this thesis there are presented two simplified and two-dimensional models. The first model assumes homogeneous heat flux in the tubes, while the other assumes homogeneous temperature. The main characteristic of the models is that they consider circumferential and axial distribution of the temperature in the receiver tubes. In addition, they take into account the main heat exchange mechanisms, as well as the temperature dependence of the thermo-mechanical properties of tube materials and heat transfer fluid. Firstly, the SPT operation modes and weakness were analysed. Subsequently, the viability of installing a system to reduce the parasitic energy consumption of the SPT was studied. This system, named Potential Energy Recovery System (PERS), recovers the potential energy from the downcomer of the receiver. The PERS was included in the models of two different actual SPT resulting in important energy savings in both plants. The simplified models were validated with CFD simulations, other simplified models, and experimental data. Regarding the CFD, the accuracy of the results is similar, but the simplified models proposed here have a significant lower computational cost, which is a notable advantage for the pre-design of the receiver where many geometrical parameters must be analysed. Regarding experimental data, given the inlet temperature of the heat transfer fluid, the direct normal irradiance, and an approximation of the aiming strategy of the heliostat field, the results obtained for the outlet temperature of the salt and the mass flow rate in the receiver are very close. Comparing with previous simplified models the thermal efficiency obtained is around 10% lower than in previous studies. The key of this difference is the thermal resistance for the heat transfer process related to the fluid and the tube material. It was also seen that the Biot number is large, and therefore the circumferential temperature must be taken into account for proper receiver efficiency estimation. In addition, different receiver geometries were analysed to find the optimum receiver design. It was determined that the most restrictive variables are the mechanical stresses and the film temperature. Regarding the receiver flow path, the best option is to implement two symmetrical paths that in the north hemisphere go from north - to - south of the receiver assuring the peak flux far from southern panels. Finally, the feasibility of employing SPT that uses supercritical or ultrasupercritical power blocks was analysed using the developed thermal models. However, the increase of the power block efficiency implies higher heat losses in the receiver. Therefore, the new generation of SPT will be only advisable when the cost of materials and systems decrease considerably.