Abstract: Disclosed herein are systems and methods for predicting localized carbon intensity of an electrical grid, comprising: (a) receiving a topography of the electrical grid; and (b) assigning one or more emissions factors to a first set of nodes or a second set of nodes associated with one or more flows of electrical power through the first set of nodes or the second set of nodes of the topography. Assigning the one or more emissions factors may comprise using a trained machine learning (ML) model to determine a power distribution at a node of the first set of nodes or the second set of nodes. The trained ML model may be trained using one or more flows of electrical power through a first set of nodes or a second set of nodes of the topography. The method may also comprise (c) determining one or more carbon intensities from the one or more emissions factors at the first set of nodes or the second set of nodes, thereby predicting the localized carbon intensity of the electrical grid.

Abstract: A system and method for forcing data center power consumption to specific levels by dynamically adjusting equipment utilization are provided.

Abstract: A system and method for using a data center as a virtual power plant are described. The data can be used to reduce energy consumption using pre-cooling and absorb excess energy generation.