Bayesian Renewables Scenario Generation via Deep Generative Networks
This provides a model-free and efficient method for scenario generation in renewable energy planning, addressing scalability issues in conventional statistical models.
The paper tackles the problem of generating diverse and accurate renewable energy scenarios by using a Bayesian GAN to capture different modes in wind and solar data, demonstrating the ability to generate clusters with varying variance and mean and distinguish mixed historical data.
We present a method to generate renewable scenarios using Bayesian probabilities by implementing the Bayesian generative adversarial network~(Bayesian GAN), which is a variant of generative adversarial networks based on two interconnected deep neural networks. By using a Bayesian formulation, generators can be constructed and trained to produce scenarios that capture different salient modes in the data, allowing for better diversity and more accurate representation of the underlying physical process. Compared to conventional statistical models that are often hard to scale or sample from, this method is model-free and can generate samples extremely efficiently. For validation, we use wind and solar times-series data from NREL integration data sets to train the Bayesian GAN. We demonstrate that proposed method is able to generate clusters of wind scenarios with different variance and mean value, and is able to distinguish and generate wind and solar scenarios simultaneously even if the historical data are intentionally mixed.