Publicação

Semi-active control of building structures using a neuro-fuzzy controller with acceleration feedback

Ver documento

Detalhes bibliográficos
Resumo:The present paper investigates the effectiveness of a neuro-fuzzy controller to reduce the response of building structures subjected to seismic excitations. The proposed controller was developed using an Adaptive Neuro-Fuzzy Inference System (ANFIS) to train a fuzzy logic system. In this case, floor acceleration measurements are used to compute the desired control signal to command a MR damper. A numerical example involving a three degrees of freedom building structure excited by the El Centro earthquake is presented to demonstrate the effectiveness of the proposed semi-active control system in reducing the response under seismic loading. A comparison between uncontrolled and controlled structural responses are used to validate the performance and efficiency of the proposed semi-active controller.
Autores principais:Braz-César, Manuel
Outros Autores:Oliveira, Kellie; Barros, Rui
Assunto:Structural control Semi-active control Neuro-fuzzy MR dampers
Ano:2015
País:Portugal
Tipo de documento:comunicação em conferência
Tipo de acesso:acesso restrito
Instituição associada:Instituto Politécnico de Bragança
Idioma:inglês
Origem:Biblioteca Digital do IPB
Descrição
Resumo:The present paper investigates the effectiveness of a neuro-fuzzy controller to reduce the response of building structures subjected to seismic excitations. The proposed controller was developed using an Adaptive Neuro-Fuzzy Inference System (ANFIS) to train a fuzzy logic system. In this case, floor acceleration measurements are used to compute the desired control signal to command a MR damper. A numerical example involving a three degrees of freedom building structure excited by the El Centro earthquake is presented to demonstrate the effectiveness of the proposed semi-active control system in reducing the response under seismic loading. A comparison between uncontrolled and controlled structural responses are used to validate the performance and efficiency of the proposed semi-active controller.