Search

International audience; 12 13 Two pilot-scale regenerative heat storage systems have been tested by the French Alternative Energies and Atomic 14 Energy Commission (CEA). The first one is a 1.1-MWhth structured packed bed consisting of ceramic plates forming 15 corrugated channels. The second one is a 1.4-MWhth granular packed bed consisting of basaltic rocks enclosed by refractory 16 walls. The two regenerators were tested over a hundred of thermal cycles between 80°C and 800°C with different fluid mass 17 flows. Both systems showed their ability to store heat efficiently and to provide thermal energy at a stable temperature for the 18 most part of the discharge process. The granular packed bed exhibited large transverse thermal heterogeneities due to flow 19 channelling in the corners of the cross section. However, this phenomenon appears not to have degraded significantly the 20 thermal performances, and the average one-dimensional thermal behaviour of the system may be assessed thanks to the surface 21 weighted average of the temperature over the bed cross section. Compared to the granular packed bed, the structured bed 22 showed comparable thermal performances while inhibiting flow heterogeneities and reducing by up to 54% the average 23 pressure drop. Furthermore, at the end of the test campaign, the packed beds were observed and compared from a mechanical 24 point of view. The thermal results were successfully simulated over numerous charge/discharge cycles thanks to a one-25 dimensional numerical model. This is significant since the discrepancies between experimental and numerical results are likely 26 to accumulate from a cycle to the other. The model considers the packed beds as continuous and homogeneous porous media 27 but takes account of the conduction resistances within the solid filler and the walls. The pressure drop of the beds was 28 computed using a correlation developed thanks to a previous CFD study for the structured packed bed, and the Ergun equation 29 for the granular packed bed. Compared to experimental data, these correlations enabled to estimate the order of magnitude and 30 the evolution trend of the pressure drop with an average deviation ranging from-7.2% to +61.9%. For the granular packed bed, 31 these deviations are ascribed to the flow heterogeneities and the shape of the rocks which are not taken into account in the 32 Ergun equation. 33 34 In order to tackle fossil fuel depletion and climate change, the renewable energy share in the energy mix has to be 40 increased, especially for electricity generation (IPCC, 2014). However, some promising renewable energy sources like wind 41 and solar are intermittent. That's why energy storage is one of the technical solutions enabling the development of a 42 continuous and controllable renewable energy supply. 43 Nowadays, electricity storage is largely dominated by Pumped Hydroelectric Energy Storage since this technology is 44 mature and offers low loss storage capacities. However, it has a very low energy and power density (Sabihuddin et al., 2015), 45 and requires specific locations with high altitude difference to be implemented. That's why this technology is unlikely to 46 respond the increasing need for large scale energy storage. Several promising alternative technologies have been proposed like 47 Adiabatic Compressed Air Energy Storage (Hartmann et al., 2012; Sciacovelli et al., 2017a), Pumped Thermal Energy Storage 48 (Desrues et al., 2010; Garvey et al., 2015), Liquid Air Energy Storage (Sciacovelli et al., 2017b) and Thermal Energy Storage 49 in Concentrated Solar Power (CSP) plants (to shift the conversion of heat into electricity). In all these thermodynamic 50 processes, there is a need for large scale Thermal Energy Storage systems. It is preferable to store/recover the energy at very 51 high or very low temperature (for Pumped Thermal Energy Storage and Liquid Air Energy Storage) to ensure good 52 thermodynamic efficiency. Regenerative sensible heat storage using gas (usually air) as heat transfer fluid is a relevant and 53 mature technology which meets this requirement. 54 It consists in storing sensible heat in a solid packed bed enclosed in an insulated tank. The packed bed may be either 55 granular or structured. The thermal energy is conveyed by a gaseous heat transfer fluid in direct contact with the solid. During 56 charging process, the hot fluid is injected by the top of the tank, heats the solid and exits at cold temperature from the bottom. 57 To discharge the system, the flow is inverted: the cold fluid is injected by the bottom, is heated up by the solid and exits at hot 58 temperature from the top. Thanks to buoyancy forces, this configuration preserves thermal stratification with hot and cold 59 regions well separated by a thermal gradient as stiff as possible. 60 This technology is already used in steel and glass industries (to preheat air in blast furnaces), and in industrial air 61 purification systems. However, to use it in the above mentioned thermodynamic processes, the design and the operation 62 strategy should be adapted to each particular case to ensure optimal performances. This optimization requires numerical 63 models validated with experimental data obtained on representative pilot-scale setups. 64