Wind farms are optimally designed to capture wind energy considering the applicability of certain wind speed and direction patterns over a long duration. With growing deviation in the weather patterns across the globe, this assumption might not be applicable after installation resulting in sub-optimal power production from the wind farm. Hence, control of yaw angles of turbines to maximize the overall power production becomes essential. In this study, Bastankhah wake effect model is used for model predictive control with an aim of maximizing the power production from a given layout. For this, the best-case and worst-case scenarios of generative adversarial networks (GAN) and data-driven robust optimization through intelligent fuzzy transcription (DRIFT) are selected. The maximum power produced for the best-case scenario with GANs and DRIFT are 5.93 GW and 7.08 GW, respectively. For the worst-case scenarios, the corresponding figures are 5.78 GW and 5.64 GW, respectively. While the central focus of this chapter is wind energy, the approach can be followed with other sources of energy like solar energy.