Generating Multivariate Load States Using a Conditional Variational Autoencoder
This addresses the problem of scenario generation for power system planning, offering an incremental improvement in generative modeling for high-dimensional dependencies.
The paper tackled the challenge of generating realistic multivariate load states for power system planning when historical data is limited, proposing a conditional variational autoencoder (CVAE) model that co-optimizes output variability to improve statistical properties, and demonstrated it outperforms other data generating mechanisms in experiments.
For planning of power systems and for the calibration of operational tools, it is essential to analyse system performance in a large range of representative scenarios. When the available historical data is limited, generative models are a promising solution, but modelling high-dimensional dependencies is challenging. In this paper, a multivariate load state generating model on the basis of a conditional variational autoencoder (CVAE) neural network is proposed. Going beyond common CVAE implementations, the model includes stochastic variation of output samples under given latent vectors and co-optimizes the parameters for this output variability. It is shown that this improves statistical properties of the generated data. The quality of generated multivariate loads is evaluated using univariate and multivariate performance metrics. A generation adequacy case study on the European network is used to illustrate model's ability to generate realistic tail distributions. The experiments demonstrate that the proposed generator outperforms other data generating mechanisms.