Réduction des pertes à vide des transformateurs de distribution par utilisation de rubans amorphes

This PhD deals with the application of amorphous ribbons in distribution transformer cores in order to reduce network losses. The material is an alloy from iron, boron and silicon compounds. The first results on the material reveal a good physicochemical compatibility with tested transformer dielectric fluids. It should be noted these materials possess a far longer life than the transformer itself. The potential for reducing losses from distribution transformers is considered as one element of EU and national strategies on energy efficiency. Losses generated through amorphous transformers are twice less than conventional ones. Amorphous ribbon units represent a significant new advance in transformer technology and losses reduction. The investment, put to purchase the product, can be easily gained by the capitalisation of losses. The return can be achieved in 10 to 12 years, depending on the purchase price. It is expected that a transformer will be subjected to a number of short-circuits during its service life, but sooner or later one such event will cause some slight winding movement, and the ability of the transformer to resist further short-circuits will then be severely reduced. Amorphous metal distribution transformers (AMDT) is no exception and they should be able to resist electrodynamic forces during short-circuit test and match ERDF specifications. In fact, during SC, extreme electrodynamic forces cause the windings to deform and this shape deformity creates a lot of shear stress on amorphous cores which lead to break-up of some ribbons. The active part of an amorphous transformer should be strong enough to resist these stresses. In fact, the behaviour of the core materials under short circuit currents depends especially on the know-how of manufacturers who can outline electrodynamic stress inside transformers. Concept design and processes have been improved in order to provide reliable devices but the task is not completely done yet. Short-circuit withstand is one of the most important aspects which will approve this technology.