The optimal planning of the annual development tasks of a medium voltage power distribution network, to tackle the evolution of the demand with maximum levels of quality and continuity of service and minimum investment, has been under the spotlight of the academic community for the last forty years. The methodologies avalaible today present some withdraws that might complicate its application on real cases. This thesis describes the design of a system of analysis and planning of real distribution networks, which incorporates a new multi-objective and dynamic multi-stage heuristic methodology to generate the optimal and efficient planning strategies. This new heuristic methodology contains strategies to increase its efficiency that have not been found in the specialized literature: the typing of the development tasks and the evaluation of the relative efficiency of the optimal planning strategies during the search. / La planificació òptima dels treballs anuals de desenvolupament d'una xarxa de distribució eléctrica de mitjana tensió, per atendre l'evolució de la demanda amb uns nivells màximes de qualitat i continuïtat del subministrament i amb una inversió mínima, ha estat objecte d'interès de la comunitat acadèmica durant els darrers quaranta anys. Les metodologies formulades fins avui presenten inconvenients que poden dificultar la seva aplicació en casos reals. Aquesta tesi detalla el disseny d'un sistema d'anàlisi i planificació de xarxes de distribució reals, que incorpora una nova metodologia heurística multi-objetiva i dinàmica multi-etapa per l'obtenció de planificacions òptimes i eficients. Aquesta nova metodologia heurística conté estratègies de millora de la seva eficiència que no s'han trobat en la literatura especialitzada: la tipificació d'actuacions de desenvolupament i l'avaluació de l'eficiència relativa de les planificacions òptimes durant el procés de cerca.

Severe accidents in light water nuclear reactors occur when reactor vessel water inventory decreases and there is no available additional water coolant to be delivered into the core. In general, during an extended severe accident sequence a period exists in which the reactor core, after a partial or total melt down, is poured into the lower plenum that can have some water present. The study of the interaction of the melt fuel with the water is the objective of MFCI (Melt Fuel Coolant Interaction) activities.MFCI is one of the most important issues awaiting resolution in water cooled reactor safety analysis. The progression of a severe accident in a water cooled reactor can lead to energetic (steam explosion) or non-energetic (melt quenching) interactions as the molten fuel relocates and eventually interacts with the coolant either in the vessel lower head (in vessel) or in the cavity (ex-vessel).The MFCI experiments at JRC Ispra site were conducted in the FARO (Furnace And Release Oven) test facility under realistic melt composition and prototypical accident conditions to provide basic information on underlying phenomena. The experimental programme was complemented by comprehensive pre-test and post-test analytical activities based on the development and application of the thermalhydraulic COMETA (COre MElt Thermalhydraulic Analysis) code. The code is developed and assessed on the basis of experimental information acquired in the FARO facility tests, and there are some limitations and uncertainties in their application to the full plant, which need to be identified and possibly quantified.In general the main objective of the PhD research was achieved expanding the general knowledge in Melt Fuel Coolant Interaction. The knowledge was complemented collaborating and complementing the application of COMETA code under conditions not experimented before, developing and improving COMETA code sources and verifying the code consistency, analysing and unifying the COMETA simulations carried so far.Also a further analytical study was carried out in order to illustrate the MFCI inside the general overview of a NPP (Nuclear Power Plant) severe accident sequence.