Optimizing the Geometry of a Half-Toroidal CVT

In this research a model is developed which calculates the torque transmission efficiency of a half-toroidal CVT as a function of variator geometry, gear ratio and input torque. The different criteria which are considered to be important in a half toroidal CVT performance are torque transmission efficiency, variator weight, roller fatigue life and bearing torque loss. Variation of geometrical characteristics such as roller curvature, number of power rollers, aspect ratio and half cone angle affect these criteria in different ways. Therefore, in order to find the optimal quantity for each geometrical parameter an optimization problem should be developed. The objective function of this optimization problem consists of inverse of torque transmission efficiency, system weight, roller fatigue life and torque loss. After choosing the appropriate weight factors, the Genetic Algorithm method is employed to minimize the objective function. It is believed that the geometry proposed in this paper can increase the efficiency of the variator to 94.4 percent.