Experiment design in compliant mechanisms and kinematic identification of parallel mechanisms

This article discusses a procedure for force-displacement modeling compliant mechanisms by using a design of computer experiments methodology -- This approach produces a force-displacement meta-model that is suited for real-time control of compliant mechanisms -- The term meta-model is used to repre...

Full description

Autores:
Restrepo Arango, David
Tipo de recurso:
Fecha de publicación:
2010
Institución:
Universidad EAFIT
Repositorio:
Repositorio EAFIT
Idioma:
spa
OAI Identifier:
oai:repository.eafit.edu.co:10784/7234
Acceso en línea:
http://hdl.handle.net/10784/7234
Palabra clave:
Metamodelos
Cinemática inversa
Ángulos de Euler
Deformación elástica
Deformaciones cuasi-estáticas
CINEMÁTICA
MOVIMIENTOS MECÁNICOS
MÉTODO DE ELEMENTOS FINITOS
DEFORMACIONES
ANÁLISIS NUMÉRICO
ALGORITMOS(COMPUTADORES)
Kinematics
Mechanical movements
Finite element method
Numerical analysis
Computer algorithms
Rights
License
Acceso abierto
Description
Summary:This article discusses a procedure for force-displacement modeling compliant mechanisms by using a design of computer experiments methodology -- This approach produces a force-displacement meta-model that is suited for real-time control of compliant mechanisms -- The term meta-model is used to represent a simplified and efficient mathematical model of unknown phenomena -- The meta-modeling of compliant mechanisms is performed from virtual experiments based on factorial- and space-filling design of experiments -- The procedure is used to model the quasi-static behavior of the HexFlex compliant mechanism -- The HexFlex is a parallel compliant mechanism for nano-manipulation that allows six degrees of freedom of its moving stage -- The meta-model of the HexFlex is calculated from experiments with the Finite Element Method (FEM) -- The obtained meta-model for the HexFlex is linear for the range of movement of the mechanism -- The accuracy of the meta-model was calculated conducting a set of computer experiments with random uniform distribution of the input forces -- Three criteria were calculated in each displacement direction (x, y, z, θx, θy, θz) comparing the meta-model prediction with respect to the results of the virtual experiments: 1. maximum of the absolute value of the error, 2. relative error, and 3. root mean square error -- The maximum errors were founded adequate with respect to demanding manufacturing tolerances (absolute errors) and lower than errors reported by other authors (relative errors)