2007
DOI: 10.1016/j.mechmachtheory.2006.07.004
|View full text |Cite
|
Sign up to set email alerts
|

Multiobjective constrained optimal synthesis of planar mechanisms using a new evolutionary algorithm

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
26
0
2

Year Published

2011
2011
2019
2019

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 66 publications
(28 citation statements)
references
References 12 publications
0
26
0
2
Order By: Relevance
“…The position error is the sum of squares of distances between each P d and the corresponding P o, where P d is a set of predefined points, and P a is a set of obtained points from an input design vector of linkage dimensions and other parameters. The members of a design solution x include r 1 , r 2 …”
Section: Optimization Problem and Constraint Handlingmentioning
confidence: 99%
See 3 more Smart Citations
“…The position error is the sum of squares of distances between each P d and the corresponding P o, where P d is a set of predefined points, and P a is a set of obtained points from an input design vector of linkage dimensions and other parameters. The members of a design solution x include r 1 , r 2 …”
Section: Optimization Problem and Constraint Handlingmentioning
confidence: 99%
“…One task for designing a four-bar linkage is a path generation problem. This synthesis problem is aimed at finding significant link lengths of the mechanism to achieve a point on a coupler link moving along the desired path [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. The problem is usually converted to be an optimization problem, which is posed to find the dimensions of a mechanism and some other parameters.…”
Section: Introductionmentioning
confidence: 99%
See 2 more Smart Citations
“…Adicional a lo antes mencionado, tal como lo señala Cabrera et al (2007), existen restricciones que deben contemplarse como: las condiciones de Grashof, que permiten una revolución completa de al menos un eslabón, la secuencia de los ángulos de entrada i 2 θ que debe ser ascendente o descendente y el especificar el rango de las variables de diseño máximo y mínimo.…”
Section: Diseño De La Función De Aptitudunclassified