Modeling and simulation of a solar cavity receiver for low latitudes

• Ano de defesa: 2020
• Autor(a) principal: Carvalho, Renan de Souza
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Cavity receiver; Central tower; Concentrated solar power; Direct steam generation; Energia solar concentrada; Geração direta de vapor; Receptor solar de cavidade; Simulação transiente; Torre central; Transient simulation.
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/106/106134/tde-30062020-150746/

Concentrated Solar Power (CSP) has been proposed as a promising renewable alternative to diversify the Brazilian electricity matrix. No CSP plants are operational in Brazil, but R&D projects are under development. In this context, the present work aimed to assess the behavior of a Direct Steam Generation (DSG) cavity receiver based on the SG4 technology to be installed in a 50 kWel central tower receiver plant, currently under construction in Caiçara do Rio do Vento, RN, Brazil, and possible improvements. The transient simulation used real data collected at the site, applying the lumped-element method for cavity spatial discretization, and Euler forward method for temporal discretization, as well as the Finite Volume Method (FVM) for the heat transfer between the steel tube and water in a mode detailed second model. Validation was carried out by comparison of results with a steady-state model considering two different days, with different irradiation values and times of the day. The results showed negligible convective heat losses compared to reflection and re-radiation losses. In addition, the receiver efficiency is higher than the receiver attached to the assessed improvements. The receiver thermal efficiency was 91.33% for the first model, and 90.18% for the detailed model, with 1,898.93 kg and 1,883.54 kg of steam generation, respectively. The validation results showed good agreement between the transient and steady-state models.