Johan Driesen

Gigawatt-hour scale savings on a budget of zero: Deep reinforcement learning based optimal control of hot water systems

Abstract 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 200xa0kWh 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.

Power electronics as the enabling technology for sustainable energy in the smart city

This paper discusses the technology trends behind the energy transition happening in the energy system of modern cities, linked to electrification, decentralization and digitalization. The role of power electronics, with a focus on building-level technologies, and the derived future requirements for converters and components are discussed. It is demonstrated using relevant use cases that power electronics represents “the new blocks that keep the building upright”.

Comparative study of current redistributor's topologies for mitigating unbalanced currents in bipolar DC microgrids

Whereas bipolar DC microgrid has several advantages over monopolar DC grid [1], [2], its disadvantages cannot be underestimated. The currents in the pole lines can become asymmetric, and the neutral conductor carries non-zero current. To eliminate the current imbalance caused by unipolar loads/generators in bipolar DC microgrids, different types of current redistributor have been proposed and implemented [3], [4]. However, the control design, switching currents and required operational power of the current redistributors have not been clearly discussed. This paper focuses on comparing single half-bridge and three phase half-bridge current redistributors by simulation and experimental validation.