Nonlinear stochastic analysis of wave energy converters via statistical linearization.

• Autor(a) principal: Silva, Leandro Souza Pinheiro da
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: http://www.teses.usp.br/teses/disponiveis/3/3135/tde-08012020-161016/
• topic: Análise estocástica; Dinâmica não linear; Domínio da frequência; Energia; Energia das ondas; Fontes alternativas de energia; Fontes renováveis de energia; Frequency domain; Linearização estatística; Nonlinear dynamics; Ondas; Statistical linearization; Stochastic analysis; Wave energy

Renewable sources of energy play a fundamental role to attend the constant rising in the global energy demand. The main objectives of utilizing renewable energies are to reduce the negative aspects associated with the utilization of fossil fuels and to diversify the energy mix. Among the renewable sources, ocean wave energy remains insufficiently explored and have the capacity to contribute to energy production. The use of wave energy has been promoted due to the vast and dense amount of energy and regularity in power distribution. The idea of harvesting wave energy exists for at least two centuries. However, it mostly started after the oil crisis of the 1970s. Since then, several wave energy converters were created without a predominant design, and more concepts are expected to be created. The challenges in the designing of wave energy converts rely on the dynamics of such systems and the stochastic nature of the environmental loads. As the device is usually set to operate near resonant conditions, wave energy devices exhibit large displacements, and nonlinear forces rise in the dynamics of the system. In this regard, the analysis of wave energy converters is usually conducted using time domain models to include the nonlinear effects. However, the computational cost associated with these simulations is high compared to traditional frequency domain models. In addition, several load conditions are necessary to evaluate the body response due to the stochastic characteristics of ocean waves, becoming undesirable to conduct time domain simulations. This thesis focuses on the stochastic analysis of wave energy converters in the frequency domain using the statistical linearization to evaluate the effects of nonlinear forces. The technique employed in this work offers a fast and reliable estimation of the device dynamics. Two conceptually different wave energy converters are investigated: a point absorber, and an oscillating water column. The results obtained using the statistical linearization are compared with their equivalent time domain models to verify the reliability of the technique. The results obtained show a good agreement was obtained between the statistical linearization and time domain simulations in terms of response distribution, power spectrum density, mean offsets, and mean power absorbed. However, the computational cost associated with time domain simulations was remarkably superior to the statistical linearization as expected. Therefore, the technique applied in this thesis offers a valuable approach to be used as a design tool for wave energy devices, and optimization procedures to develop the wave energy sector.