The primary objective of this paper is to demonstrate improved energy efficiency for domestic hot water (DHW) production in residential buildings. This is done by deriving data-driven optimal heating schedules (used interchangeably with policies) automatically. The optimization leverages actively learnt occupant behaviour and models for thermodynamics of the storage vessel to operate the heating mechanism -an air-source heat pump (ASHP) in this case -at the highest possible efficiency. The proposed algorithm, while tested on an ASHP, is essentially decoupled from the heating mechanism making it sufficiently robust to generalize to other types of heating mechanisms as well. Simulation results for this optimization based on data from 46 Net-Zero Energy Buildings (NZEB) in the Netherlands are presented. These show a reduction of energy consumption for DHW by 20% using a computationally inexpensive heuristic approach, and 27% when using a more intensive hybrid ant colony optimization based method. The energy savings are strongly dependent on occupant comfort. This is demonstrated in real world settings for a low-consumption house where active control was performed using heuristics for 3.5 months and resulted in energy savings of 27% (61 kWh). It is straightforward to extend the same models to perform automatic demand side management (ADSM) by treating the DHW vessel as a flexibility bearing device.
Energy consumption for hot water production is a major draw in high efficiency buildings. Optimizing this has typically been approached from a thermodynamics perspective, decoupled from occupant influence. Furthermore, optimization usually presupposes existence of a detailed dynamics model for the hot water system. These assumptions lead to suboptimal energy efficiency in the real world. In this paper, we present a novel reinforcement learning based methodology which optimizes hot water production. The proposed methodology is completely generalizable, and does not require an offline step or human domain knowledge to build a model for the hot water vessel or the heating element. Occupant preferences too are learnt on the fly. The proposed system is applied to a set of 32 houses in the Netherlands where it reduces energy consumption for hot water production by roughly 20% with no loss of occupant comfort. Extrapolating, this translates to absolute savings of roughly 200 kWh for a single household on an annual basis. This performance can be replicated to any domestic hot water system and optimization objective, given that the fairly minimal requirements on sensor data are met. With millions of hot water systems operational worldwide, the proposed framework has the potential to reduce energy consumption in existing and new systems on a multi Gigawatt-hour scale in the years to come.
Abstract. This paper aims to design an algorithm dedicated to operational planning for microgrids in the challenging case where the scenarios of production and consumption are not known in advance. Using expert knowledge obtained from solving a family of linear programs, we build a learning set for training a decision-making agent. The empirical performances in terms of Levelized Energy Cost (LEC) of the obtained agent are compared to the expert performances obtained in the case where the scenarios are known in advance. Preliminary results are promising.
The Mas2tering and DRIvE European projects develop a software platform to manage the residential and tertiary energy flexibility in local communities of prosumers. This platform includes forecasting, optimization, cybersecurity and fast-response capabilities modules. Business models and use cases were developed, taking into account the regulatory challenges. The platform will be implemented in five pilot sites across three countries.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.