Generating synthetic scenarios for time-series data is crucial for various applications, including predictive modeling, data augmentation, and scenario analysis. In this study, we focus on creating scenarios in purpose of data augmentation for the optimal design of renewable energy systems within microgrids by testing two methods: directly generating photovoltaic production data and generating meteorological data scenarios followed by calculating the production. Our goal is to evaluate how well the TimeGAN artificial intelligence tool performs in generating realistic synthetic scenarios for both approaches. TimeGAN, a type of Generative Adversarial Network (GAN), is known for its ability to capture temporal patterns and maintain data distribution. We use TimeGAN to generate synthetic scenarios for renewable energy production and compare the feasibility and quality of the scenarios produced by each method. Our comparative analysis highlights the benefits and limitations of both approaches, offering valuable insights into scenario generation in energy systems.