Identificação de sistemas modelo Narx para a fadiga muscular emindivíduos sem deficiência durante contrações isométricas voluntárias

Detalhes bibliográficos
Ano de defesa: 2018
Autor(a) principal: Amanda Fernandes Vilaça Martins
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal de Minas Gerais
UFMG
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://hdl.handle.net/1843/RAOA-BEKLW4
Resumo: This work consists of the development of a system identification methodology capable of modeling muscle fatigue during voluntary isometric movements of individuals without physical disability, using surface electromyographic (sEMG) and torque data. A protocol was performed with the dominant lower limb of the volunteer so that muscle fatigue was achieved. The torque measurement was performed by the isokinetic Biodex System 4 Pro and EMG was performed by the EMG_800C with non - invasive surface Ag / AgCl electrodes and in the bipolar configuration. Labview software was used for the acquisition of this data and Matlab® for signal processing and identification of the system studied. The objective was to find a model that relates two measures of muscular fatigue, using for this the electromyography as input and the torque as output of the system. In this analysis, the proposed methodology involved the NARX (Nonlinear Auto-Regressive with eXogenous Inputs) model, with the error reduction rate (ERR), Akaike (AIC) and Bayes (BIC) criteria algorithms. The results show that the model proposed for each volunteer can be used due to the statistical similarity found. In addition, all the suggested models obtained good results when compared to other studies in the literature. The mean RMS error for the model with 15 regressors was 5.24% ± 2.59%; already for 10 regressors was 6.89% ± 2.83% and; for the general model, 5.54% ± 3.21. This allows concluding that the methodology adopted in this workwas successful for the conditions studied.