Selective Exhaust Gas Recirculation (S-EGR) consists of selectively transferring CO2 from the exhaust gas stream of a gas-fired power plant into the air stream entering the gas turbine compressor. Unlike in “non-selective” Exhaust Gas Recirculation (EGR) technology, recirculation of, principally, nitrogen does not occur, and the gas turbine still operates with a large excess of air. Two configurations are proposed: one with the CO2 transfer system operating in parallel to the post-combustion carbon capture (PCC) unit; the other with the CO2 transfer system operating downstream of, and in series to, the PCC unit. S-EGR allows for higher CO2 concentrations in the flue gas of approximately 13-14 vol%, compared to 6.6 vol% with EGR at 35% recirculation ratio. The oxygen levels in the combustor are approximately 19 vol%, well above the minimum limit of 16 vol% with 35% EGR reported in literature. At these operating conditions, process model simulations show that the current class of gas turbine engines can operate without a significant deviation in the compressor and the turbine performance from the design conditions. Compressor inlet temperature and CO2 concentration in the working fluid are critical parameters in the assessment of the effect on the gas turbine net power output and efficiency. A higher turbine exhaust temperature allows the generation of additional steam which results in a marginal increase in the combined cycle net power output of 5% and 2% in the investigated configurations with S-EGR in parallel and S-EGR in series, respectively. With aqueous monoethanolamine scrubbing technology, S-EGR leads to operation and cost benefits. S-EGR in parallel operating at 70% recirculation, 97% selective CO2 transfer efficiency and 96% PCC efficiency results in a reduction of 46% in packing volume and 5% in specific reboiler duty, compared to air-based combustion CCGT with PCC, and of 10% in packing volume and 2% in specific reboiler duty, compared to 35% EGR. S-EGR in series operating at 95% selective CO2 transfer efficiency and 32% PCC efficiency results in a reduction of 64% in packing volume and 7% in specific reboiler duty, compared to air-based, and of 40% in packing volume and 4% in specific reboiler duty, compared to 35% EGR. An analysis of key performance indicators for selective CO2 transfer proposes physical adsorption in rotary wheel systems as an alternative to selective CO2 membrane systems. A conceptual design assessment with two commercially available adsorbent materials, activated carbon and Zeolite X13, shows that it is possible to regenerate the adsorbent with air at near ambient temperature and pressure. Yet, a significant step change in adsorbent materials is necessary to design rotary adsorption systems with dimensions comparable to the largest rotary gas/gas heat exchanger used in coal-fired power plants, i.e. approximately 24 m diameter and 2 m height. An optimisation study provides guidelines on the equilibrium parameters for the development of materials. Finally, a technical feasibility study of configuration options with rotary gas/gas heat exchangers shows that cooling water demand around the post-combustion CO2 capture system can be drastically reduced using dry cooling systems where gas/gas heat exchangers use ambient air as the cooling fluid. Hybrid cooling configurations reduce cooling and process water demand in the direct contact cooler of a wet cooling system by 67% and 35% respectively, and dry cooling configurations eliminate the use of process and cooling water and achieve adequate gas temperature entering the absorber.

Cette thèse s'intéresse à la coordination des sous-systèmes d'un nouveau genre de centrale de production d'énergie : un cycle combiné hybride (HCC - Hybrid Combined Cycle). Cette centrale HCC n'existe pas encore mais combine un cycle combiné gaz (CCG), un moyen de production solaire thermodynamique (miroirs cylindro-paraboliques) et un moyen de stockage thermique (stockage indirect de chaleur sensible utilisant deux réservoirs de sels fondus). Comment coordonner ces trois sous-systèmes de manière optimale lors des variations de demande de puissance ou des prix du gaz ?Dans un premier temps, chacun des trois sous-systèmes est étudié de manière indépendante afin d'obtenir, d'une part, un modèle physique permettant de caractériser le comportement dynamique du sous-système considéré et, d'autre part, un contrôle local qui agit en fonction des objectifs de fonctionnement prédéfinis. Un modèle du système complet interconnecté de l'HCC est ensuite obtenu en couplant les modèles des trois sous-systèmes. Enfin, une coordination des différents sous-systèmes est mise en place pour adapter le fonctionnement de chacun, en fonction des objectifs globaux de la centrale HCC complète, en optimisant les consignes de chaque sous-système. Dans ce travail, une coordination de type linéaire quadratique et une coordination de type optimale prédictive sont étudiées. Les résultats obtenus sont bien prometteurs : ils montrent, tout d'abord, que lors d'un appel de puissance, la commande coordonnée permet au système HCC de répondre plus rapidement, en utilisant plus efficacement la partie solaire. De plus, lorsque la demande subit beaucoup de variations, la partie solaire et la partie stockage absorbent toutes ces variations et la Turbine à Combustion (TAC) du CCG est beaucoup moins sollicitée. Lorsqu'il n'y a plus d'irradiation solaire, la partie stockage prend la relève pour continuer à produire de la vapeur solaire, jusqu'à ce que les stocks se vident. Finalement, le stockage permet d'ajuster la production de la TAC en fonction des prix du gaz. / This work concerns the subsystems coordination of a new type of power plant: a Hybrid Combined Cycle (HCC). This HCC plant is not yet build but consists of a Combined Cycle Power Plant (CCPP), a concentrated solar plant (parabolic trough) and a thermal storage system (a molten-salts two-tank indirect sensible thermal storage). How to coordinate these three subsystems optimally during variations in power demand or in gas price?First, each subsystem is studied independently in order to get on one hand a physical model that reproduces the dynamical behavior of the considered subsystem, and on the other hand, a local control that achieves an operation according to pre-specified objectives. Then, a model of the HCC system is obtained by coupling the models of the three defined subsystems.Eventually, a coordination of the subsystems is set up in order to adapt the behavior of each subsystem according to the global objectives for the full HCC system, by optimizing subsystem setpoints. In this study, a linear quadratic coordination and a model predictive coordination are designed. The obtained results are promising: they first show that during a power demand, the coordination allows the global system to quickly respond, using extensively the solar production. Besides, when the power demand undergoes many fluctuations, the solar and storage parts absorb these variations and the gas turbine of the CCPP is much less stressed. In addition, when there is no more solar radiation, the storage part continues producing solar steam, until storage tanks are empty. At last, the storage part allows to adjust the gas turbine production according to the gas prices.