Contextual Quality-Diversity Evolutionary Reinforcement Learning for HVAC Control in Tropical Commercial Buildings
This paper presents a contextual quality-diversity evolutionary reinforce learning controller, CQD-ERL, for supervisory control of a tropical chiller plant, maintaining a product archive of specialized policies and using a safety shield.
Proposing a new contextual quality-diversity evolutionary reinforce learning controller for supervisory control of a tropical chiller plant.
Before reading this…
Applications
- →Supervisory control of tropical chiller plants
To understand this paper, make sure you know these concepts first:
- Reinforce learning, evolutionary algorithms, supervisory control, chiller plantsfind papers →
Abstract
More Like ThisThis paper proposes a contextual quality-diversity evolutionary reinforcement-learning controller, CQD-ERL, for the supervisory control of a tropical, water-cooled chiller plant and its associated air side. Rather than converging to a single scalarised policy, the controller maintains a product archive of specialised policies indexed jointly by a data- driven operating context, a cluster of daily weather and load regime, and a context-invariant behaviour descriptor, filled by a gradient-free evolutionary operator and a soft-actor-critic policy-gradient operator that share one replay buffer. Every action is filtered through a deterministic safety shield before execution. The controller is trained on a two-tier reduced-order environment representing the latent load, cooling-tower approach and humidity constraints of a Singapore commercial building, and is evaluated over a full annual backtest against an ASHRAE Guideline 36 baseline.