Dimensional synthesis of planar mechanisms using neural networks: application to path generator linkages

We propose an original method to synthesize the dimensions of a planar mech anism (linkage) whose function is to generate a trajectory shape. Most grap hical and analytical synthesis methods for path generators require specifyi ng the desired trajectory in a non-functional manner, by a list of points r ather than a pure shape. Concerning the dimensional optimization methods, t hey turn out to be slow and their convergence depends on the initial soluti on. Alternatively, we propose a case-based approach (i.e., couples of traje ctories and dimensions of a given structure mechanism) using a neural netwo rk. The first stage consists in the generation of a huge case number throug h kinematic simulations, for random values of dimensions, and in a learning process of the neural network. In the second stage, of utilization, the ne ural network instantaneously makes it possible to obtain an approximate sol ution of the synthesis problem, which is an interpolation of close cases. W e show on the four-bar linkage example the good quality of the synthesized solutions, for a tiny size of the network. Next, these solutions may be use d as judicious initial solutions for a conventional dimensional optimizatio n. (C) 2001 Elsevier Science Ltd. All rights reserved.