Structural hierarchical learning for energy networks

TL;DR

This paper introduces structurally-informed neural network designs for hierarchical learning in energy networks, improving forecast accuracy and coherency, especially in data-limited scenarios, by leveraging hierarchy topologies.

Contribution

It proposes novel neural network architectures inspired by hierarchy structures, enhancing data efficiency and coherency in hierarchical forecasting methods.

Findings

Structural models with fewer connections perform best in data-limited settings.

Coherency information improves forecast accuracy and coherency when individual forecasts are reasonably accurate.

The proposed methods are resource-efficient and enhance hierarchical learning performance.

Abstract

Many sectors nowadays require accurate and coherent predictions across their organization to effectively operate. Otherwise, decision-makers would be planning using disparate views of the future, resulting in inconsistent decisions across their sectors. To secure coherency across hierarchies, recent research has put forward hierarchical learning, a coherency-informed hierarchical regressor leveraging the power of machine learning thanks to a custom loss function founded on optimal reconciliation methods. While promising potentials were outlined, results exhibited discordant performances in which coherency information only improved hierarchical forecasts in one setting. This work proposes to tackle these obstacles by investigating custom neural network designs inspired by the topological structures of hierarchies. Results unveil that, in a data-limited setting, structural models with fewer connections perform overall best and demonstrate the coherency information value for both accuracy and coherency forecasting performances, provided individual forecasts were generated within reasonable accuracy limits. Overall, this work expands and improves hierarchical learning methods thanks to a structurally-scaled learning mechanism extension coupled with tailored network designs, producing a resourceful, data-efficient, and information-rich learning process.