Multi-objective Optimization under Uncertainty of a Continuously Variable Transmission for a 1.5 MW Wind Turbine

This article presents an original methodology for the multi-objective optimization of Continuously Variable Transmission (CVT) for a wind turbine (WT). The objective functions of this optimization problem are to minimize the weight and maximize efficiency. This methodology also considers the variations of parameters caused by different factors (manufacturing tolerance, uncertainties in the operating conditions). Using a probabilistic model, the proposed algorithm combines a propagation of uncertainties and an optimization of the function objectives. The optimization is performed using the Non-dominated Sorting Genetic Algorithm (NSGA-II) with the advantage of exploring the global design space and finding the best compromise between the objectives. In order to verify the solution obtained by this approach, results were compared to the ones obtained by a previous study.