Process for optimization based on inductive approach in design of a hydraulic transmission

Abstract

This study has proposed the optimization process regarding the hydraulic transmission installed with construction and agricultural vehicles, in order to contribute to the quietness of the vehicles in which noise from the hydraulic system is expected to become noticeable due to the diversification of power sources. The pressure ripple as the noise source which is derived from the relative motion of pistons, cylinders and valves is non-linear system, and the genetic algorithm which is one of the evolutionary computations is used to search for the optimum solution of design variables. In addition to just search for the optimum solution, the readability of the design space has been obtained by the inductive approach, using many datasets generated in the process of the evolutionary computation. The decision tree, which is one of the techniques in machine learning, is used for the inductive approach and can make the readability of the design space regarding the pressure ripple generated in the hydraulic transmission without the dimensional reduction of the design space and the neglected interaction.