Agora exibindo 1 - 2 de 2
- Capítulo de livroA Study on Efficient Reinforcement Learning Through Knowledge Transfer(2022-10-05) GLATT, R.; DA SILVA, F. L.; Reinaldo Bianchi; COSTA, A. H. R.© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.Although Reinforcement Learning (RL) algorithms have made impressive progress in learning complex tasks over the past years, there are still prevailing short-comings and challenges. Specifically, the sample-inefficiency and limited adaptation across tasks often make classic RL techniques impractical for real-world applications despite the gained representational power when combining deep neural networks with RL, known as Deep Reinforcement Learning (DRL). Recently, a number of approaches to address those issues have emerged. Many of those solutions are based on smart DRL architectures that enhance single task algorithms with the capability to share knowledge between agents and across tasks by introducing Transfer Learning (TL) capabilities. This survey addresses strategies of knowledge transfer from simple parameter sharing to privacy preserving federated learning and aims at providing a general overview of the field of TL in the DRL domain, establishes a classification framework, and briefly describes representative works in the area.
- Capítulo de livroInfluence of turbine inlet temperature on the efficiency of externally fired gas turbines(2014-01-01) DE MELLO, P. E. B.; SCUOTTO, S.; DOS SANTOS, O. F.; Gustavo Donato© Springer International Publishing Switzerland 2014.Many researchers have considered externally fired gas turbines (EFGT) as an option for the implementation of biomass-fueled power plants. The EFGT cycle with regeneration or the gas-vapor combined cycle using one EFGT, also known as externally fired combined cycle (EFCC), could lead to significant efficiency improvements if compared to current technology used for power generation from biomass. This work presents one improved numerical model used for the simulation of EFGT cycle. The results were obtained with a numerical model for the EFGT cycle coupled with a model for the high temperature heat exchanger (HTHE) that is necessary for the cycle implementation. The model of the heat exchanger is based in correlations for the Colburn and friction factors, obtained with CFD simulations. In previous work, the model included only laminar regime for the heat exchanger. The present work extends the correlations that describe the behavior of the heat exchanger to turbulent and transitional regimes. The updated model of the EFGT cycle is used to investigate the influence of the turbine inlet temperature over the cycle efficiency. The results obtained confirm that the pressure drop caused by the heat exchanger is one important parameter that influences the cycle efficiency. The feasibility of the EFGT cycle is discussed taking into consideration that the highest temperature in EFGT cycle is not in the turbine inlet, but in the high temperature heat exchanger.