Reviewed by Lexie CornerMar 18 2025
A study published in Sustainable Cities and Society details the development of a predictive control system by Empa researchers. This system optimizes energy management at the building level while ensuring user comfort.
The UMAR unit at NEST served as a testbed for the experimental study. Image Credit: Zooey Braun
More renewable energy alone is not sufficient to ensure a reliable energy supply in the future; advanced management systems are needed to efficiently handle production, distribution, and consumption.
As the transition to renewable energy progresses, buildings play an increasingly vital role in the development of sustainable energy systems. While photovoltaic systems have shown promise on a small scale, such as for single-family homes, concerns remain regarding the overall security of the energy supply.
The key question is whether renewable energy sources can consistently provide enough power year-round, or whether issues like the energy scarcity experienced in winter 2023 will continue. However, the primary challenge lies not in generating renewable energy, but in the logistics of distribution.
Historically, the focus has been on power plants that provide a consistent amount of energy to the electricity grid. To meet future energy demands from renewable sources, two factors must be addressed: an increase in production facilities and the deployment of smart technologies to ensure grid stability.
Unlike traditional energy sources such as coal or uranium, solar plants do not always produce the same amount of electricity. Production depends on weather conditions and the day-night cycle. Therefore, energy consumption must be reduced during low-production periods, such as at night, and production peaks must be managed to prevent overloading the grid.
Automated systems can effectively manage these complex logistics. By factoring in local production, storage capacity, and grid availability, these systems can optimize electricity usage, ensuring grid stability and consumer flexibility.
In practical terms, predictive energy planning enables buildings to function normally, such as allowing hot showers or cooking, even when electricity production does not meet demand. Simultaneously, excess energy is not always stored locally but is fed into the grid to ensure continuous supply.
From Theory to Practice: The Test at NEST
To show the potential of such automated systems, Empa’s Urban Energy Systems Lab researchers conducted a field experiment at NEST to determine the extent to which an inhabited building may incorporate many variable demand criteria under one roof. They were primarily concerned with lowering carbon emissions, increasing energy demand flexibility, and improving resident comfort.
Using a predictive control algorithm, the team improved energy management within the building with the following configuration: a photovoltaic system for electricity production, a battery storage system, a heat pump, and a bidirectional charging station for e-vehicles. The major purpose was to reduce carbon emissions during operation by pulling electricity from the grid whenever it was available from renewable sources.
Furthermore, minimum temperatures were set for the interior rooms and the hot water tank, and the algorithm prepared customer flexibility for improved grid functioning.
The Building as an Actor in the Energy System
To demonstrate the potential of automated systems, researchers from Empa's Urban Energy Systems Lab conducted a field experiment at NEST to assess how a building could incorporate various demand criteria. Their primary goals were to reduce carbon emissions, enhance energy demand flexibility, and improve resident comfort.
Using a predictive control algorithm, the team successfully optimized energy management within the building. The configuration included a photovoltaic system for electricity generation, a battery storage system, a heat pump, and a bidirectional charging station for electric vehicles. The main objective was to minimize carbon emissions during operation by drawing electricity from the grid when renewable energy sources were available.
In addition, the team set minimum temperature levels for interior rooms and the hot water tank, and the algorithm prepared for customer flexibility to improve grid performance.
Through a Start-Up into the Market: Technology Transfer “Made by Empa”
Cai and his colleague Federica Bellizio are working to bring their technology to market through their start-up, "Kuafu." Bellizio recently received the "Empa Entrepreneur Fellowship," a scholarship for researchers aiming to establish their own businesses.
They aim to use their data-driven technology to connect grid operators and energy providers, contributing to energy optimization and decarbonization in the building and electric mobility sectors.
Journal Reference:
Cai, H. and Heer, P. (2025) Experimental implementation of an emission-aware prosumer with online flexibility quantification and provision. Sustainable Cities and Society. doi.org/10.1016/j.scs.2024.105531