Modal analysis of a parallel kinematic machine: Freehex.
Parallel Kinematic Machines (PKM) are mechanical systems utilized to produce precise operations at high speeds compared to serial machines. Its applications cover a wide range of operations in the automotive and aerospace industry. These machines can be affected by oscillatory effects such as reson...
| Autor principal: | |
|---|---|
| Formato: | Tesis |
| Lenguaje: | inglés |
| Publicado: |
2017
|
| Acceso en línea: | http://eprints.uanl.mx/14411/1/1080252113.pdf |
| Sumario: | Parallel Kinematic Machines (PKM) are mechanical systems utilized to produce precise operations at high speeds compared to serial machines. Its applications
cover a wide range of operations in the automotive and aerospace industry. These machines can be affected by oscillatory effects such as resonant frequencies, which
can produce adverse effects on the performance and the useful life of the systems.
Therefore the determination of these characteristics is an important element in the design and operation of these machines.
The research developed in this thesis addresses the identification of these resonant frequencies in the possible working volumes of a PKM called FreeHex. This
machine is used to perform high speed machining operations and in situ rapair of aerospace equipment. Two methods are developed and described to estimate the
resonant frequencies.The first is formulated through the Finite Element Method (FEM), and given the initial configuration of the FreeHex is capable to estimate
the resonant frequencies within the working volume of that configuration. The second method is intended to validate the model described by the FEM. As the
FreeHex is able to achieve a several configurations the analysis is performed to a single one and then the experimental and the finite element analysis model
are compared. The experimental results are obtained by impact testing trough Multiple Input Single Output (MISO) method on the structure. The average error between the FEA models and experimental results lies in a value of 3.94 % showing a good correlation between both methods and making possible to implement the algorithms here developed. |
|---|