Optimal-robust design of smart structures to enhance the performances of fatigue life and vibration energy harvesting

• Ano de defesa: 2024
• Autor(a) principal: Sena, João Pedro
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Mecânica
• Palavras-chave em Português: Smart structures; Harvesting; Fatigue; Shunt circuit; Topology optimization; Estruturas inteligentes; Fadiga; Circuito shunt; Otimização topológica; CNPQ::ENGENHARIAS::ENGENHARIA MECANICA::MECANICA DOS SOLIDOS::DINAMICA DOS CORPOS RIGIDOS, ELASTICOS E PLASTICOS; Engenharia mecânica; Otimização estrutural; Espaços topológicos; Energia - Consumo; ODS::ODS 9. Indústria, Inovação e infraestrutura - Construir infraestrutura resiliente, promover a industrialização inclusiva e sustentável, e fomentar a inovação.
• Link de acesso: https://repositorio.ufu.br/handle/123456789/43183; https://doi.org/10.14393/ufu.te.2024.540

Fatigue analysis in structures subjected to random loads is more efficient in the frequency domain due to lower computational cost, considering the gains from using methods such as the Ritz modal basis and neural network metamodeling. Among vibration control methods, smart materials, especially piezoelectrics, have stood out due to their energy harvesting capabilities, functioning as both sensors and actuators, and being lightweight and easy to handle. The association of shunt circuits with PZT material is achieved by tuning the circuit elements, which requires an optimization process to obtain the optimal parameters. This work proposes a robust-optimal methodology for designing dynamic systems using passive control to reduce structural fatigue and increase harvested energy. Four main cases were analyzed: full-layer circuit optimization; partial treatment circuit optimization; topological optimization by patch location; and topological optimization by PZT layer density. The optimization techniques employed, considering uncertainties in circuit parameters, proved effective in achieving better results compared to traditional methods.