The synthesis of hybrid mechanisms using genetic algorithms

• Main Author: Connor, Andrew Miles
• Published: Liverpool John Moores University 1996
• Subjects: 621.8; TJ Mechanical engineering and machinery
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.361515

This thesis presents a novel design methodology for the synthesis of hybrid mechanisms using Genetic Algorithms. GAs are a search and optimisation method which model the mechanics of population genetics to give a truly global search method. In parallel to the development of a suitable GA, the work also develops novel objective function criteria which go some way to providing an approximation to dynamic criteria whilst using only kinematic properties during calculations. This has considerable effect in reducing the time required to find a feasible solution. The thesis presents a set of results which validate the proposed methodology, both in terms of speed of convergence and quality of the final solutions obtained. The application chosen is the synthesis of a hybrid five bar path generating mechanism. A description is given of the development of a practical machine for a given test case, so as to illustrate that the solutions produced are feasible in terms of real world implementation. Results are presented which show the effectiveness of the machine. Finally, a critical analysis of both the methodology and the results is carried out. This highlights some areas in which the methodology could be improved by future work.