Transactive management of flexibility markets through hierarchical clustering

With the massive integration of renewables, consumers are becoming key system actors by offering flexibility to the grid. Moving from centralized control of a limited number of entities to managing an ecosystem of countless active resources — inherently heterogeneous 'flexumers' — poses an unprecedented scale challenge: centralized approaches suffer from high computational cost, while decentralized strategies risk neglecting network topology.

This doctoral project develops a transactive flexibility management method that overcomes this scale challenge through adaptive hierarchical spectral clustering of flexibility providers, from the market operator's perspective: simplifying market interactions and improving feasibility without significantly reducing economic efficiency.

Main objective

Develop an adaptive clustering algorithm for flexibility providers, from the market operator's perspective, to improve market reliability, settlement feasibility and system adaptability.

Methodology

Four stages: formulation of a local flexibility market (social-welfare optimization); spectral clustering via electrical, geographic and behavioral affinity aggregation; adaptive mechanisms and hierarchical settlement; validation via Python simulations and Pareto-front analysis.