Comparison of Strategies for the Optimization/Innovation of Crankshaft Balance
Engine crankshafts are required to be balanced. The balance of a crankshaft is one of several parameters to be analyzed during the design of an engine, but certainly a poor balance leads to a low life time of the whole system. It is possible to optimize the balance of a crankshaft using CAD and CAE software, thanks to the new optimization tools based on Genetic Algorithms (GA) and tools for the integration of the CAD-CAE software. GAs have been used in various applications, one of which is the optimization of geometric shapes, a relatively recent area with high research potential. This paper describes a general strategy to optimize the balance of a crankshaft. A comparison is made among different tools used for the sustaining of this strategy. This paper is an extension of a previous paper by the authors  but now different tools are being included to improve the performance of the strategy. The analyzed crankshaft is modeled in commercial 3D parametric software. A Java interface included in the CAD software is used for evaluating the fitness function (the balance). Two GAs from different sources and platforms are used and then they are compared and discussed.
KeywordsGenetic Algorithm Fitness Function Multiobjective Optimization Pareto Frontier Multi Objective Genetic Algorithm
- 1.Aguayo, Humberto, and Noel Leon. 2006. Computer Aided Innovation of Crankshafts Using Genetic Algorithms. Chap. in Knowledge Enterprise: Intelligent Strategies in Product Design, Manufacturing, and Management. IFIP International Federation for Information Processing. 471–476. Springer Boston.Google Scholar
- 2.Eldred, Michael, Anthony Guinta, William Hart, John Eddy, and Josh Griffin, Technical Report. October 2006. DAKOTA Version 4.0 User’s Manual. Albuquerque, NM, USA: Sandia National Laboratories.Google Scholar
- 3.Valenzuela-Rendón, Manuel. The Virtual Gene Genetic Algorithm. Chap. in Genetic and Evolutionary Computation GECCO 2003.Google Scholar
- 5.Coello Coello, Carlos Artemio. 1996. An Empirical Study of Evolutionary Techniques for Multiobjective Optimization in Engineering Design. Doctor of Philosophy diss., Tulane University.Google Scholar
- 6.Cueva de Leon, Jose Maria. 2006. Automatic shape variations for optimization purposes. Master in Sciences thesis, Technologico de Monterrey (ITESM).Google Scholar