Transfer of knowledge to industry
Département des Technologies des NanoMatériaux (LITEN)
Laboratoire de Formulation des Matériaux
01-11-2020
SL-DRT-20-0329
Energy efficiency for smart buildings, electrical mobility and industrial processes (.pdf)
Due to their remarkable magnetic properties, permanent magnets made of NdFeB alloys are an important part of the Energy Transition, with several applications in Energy (wind turbines) and in Transport (electric vehicle) sectors, for example. NdFeB magnets are usually produced by powder compression and sintering, and complex shapes are obtained through expensive machining operations. The powder injection moulding (PIM) process allows the direct production of parts with complex geometries through the conventional plastics processing technics, and is a way to reduce both machining operations and waste materials. Therefore, PIM is currently under consideration for the manufacturing of permanent NdFeB magnets with high density and magnetic performances and complex geometries. Nevertheless, the use of organic polymer binders for injection moulding (i), and the post-injection chemical and physical debinding steps (ii) during the PIM procedure, can be responsible of potential organic (i.e. carbon and/or oxygen) contaminations of the NdFeB powder, and consequently, of a significant degradation of magnetic properties of the magnets. Each of these contributions needs to be in-depth studied, for optimizing the magnetic properties of injection-moulded permanent NdFeB magnets. In particular, the understanding of the physicochemical interactions between polymer binders (and/or their degradation products) with the NdFeB powders, should lead to the development of feedstocks compatible with the injection moulding of low-contaminated permanent magnets.
Département des Technologies Solaires (LITEN)
Laboratoire Systèmes Electriques Intelligents
01-10-2020
SL-DRT-20-1068
Energy efficiency for smart buildings, electrical mobility and industrial processes (.pdf)
In a context of massive integration of non-dispatchable renewable production (wind and photovoltaic), the fullfilment of « production = consumption » will imply in the future to act more and more on the second term of this equality. Many questions arise about the real potential of consumption flexibility in a smartgrid, and this depending on the geographic level and also the time range of activation. The management of flexibilities as well as their valuation, will require various models, different depending on the markets (ancillary services, SPOT, intraday, balancing market) and the geographic level considered (house, building, district, city). The thesis will focus on developing: - Electrical consumption and flexibilities modelling - Learning algorithms / parameterization of these models. These will be oriented ?big data?. - Methodologies for getting one model from another To carry out this work, we will use existing simulation tools for the finest spatial scales as well as a consumption measurement database (« Linky » smart metering).
