A new predictive control strategy could help building operators reduce energy costs and improve the use of renewable power without the high computing requirements associated with more complex optimisation systems.
The study, titled A predictive rule based control strategy for integrated building energy systems with photovoltaic, heat pump, thermal tank and electric vehicles, examines the operation of buildings equipped with photovoltaic systems, heat pumps, thermal storage tanks and electric vehicles.
Buildings account for about 30% of global energy consumption, making their energy management an important part of efforts to improve efficiency and reduce emissions. Integrated building energy systems can increase renewable energy use, lower operating costs and provide greater flexibility to the electricity grid, but these benefits depend heavily on how the equipment is controlled.
Conventional rule based control is widely used because it is relatively simple and easy to implement. However, it generally relies on fixed operating rules and has limited ability to respond to future changes in electricity demand, solar generation, weather conditions and electricity tariffs.
Model predictive control can address this limitation by using forecasts to optimise equipment operation over a future period. Its drawback is the substantial computing power and complex online optimisation required, particularly when several devices and storage systems operate together.
The researchers developed a predictive rule based control strategy designed to combine the practical simplicity of conventional rule based control with the forecasting capability of model predictive control.
The strategy uses forecasts for weather, building demand and photovoltaic generation to determine the most suitable charging and discharging power, as well as target energy levels for the thermal tank and electric vehicles during different tariff periods. It then determines the actual operation of the heat pump, storage systems and other equipment based on real time system conditions and the overall energy balance.
Results
The results indicate that the proposed strategy delivered performance close to model predictive control while requiring only slightly more computing time than conventional rule based control.
During the one week simulation, the average computing time for each control action was 0.18 seconds for conventional rule based control, 0.27 seconds for the predictive rule based strategy and 239.81 seconds for model predictive control.
Total operating costs were CNY 3,446.93 under conventional rule based control. The predictive rule based strategy reduced this to CNY 2,176.37, while model predictive control achieved CNY 2,034.95.
The results show that the predictive strategy captured most of the cost saving available through model predictive control, while avoiding its significantly higher computational burden.
The strategy also reduced reliance on grid electricity during weekday peak periods. Grid purchases represented about 70% of total electric load under conventional rule based control, compared with approximately 20% under the predictive strategy and 12% under model predictive control.
Potential for practical deployment
According to the study, the results demonstrate that predictive rule based control can continuously adjust the operation of multiple storage devices according to changing conditions, rather than relying only on fixed operating modes, setpoints or thresholds.
This is particularly relevant for buildings with electric vehicles, whose availability and charging schedules can vary throughout the day. Coordinating electric vehicle charging with photovoltaic output, building demand, electricity prices and thermal storage could help reduce peak demand and improve the use of locally generated solar power.
The researchers conclude that the proposed strategy provides a lower complexity alternative for integrated building energy systems. Its combination of predictive capability, reduced computing requirements and relatively strong economic performance could support practical applications where full model predictive control is difficult or costly to implement.
The study provides a potential foundation for deploying more flexible energy management systems in commercial and residential buildings, including projects seeking to integrate distributed solar generation, efficient heating technologies, thermal storage and electric mobility.
Link to the full paper HERE 
Author: Bryan Groenendaal






Δ

August 10, 2026
August 17, 2026
September 5, 2026
August 20, 2026
August 13, 2026
September 21, 2026
Disclaimer | Privacy Policy | Terms & Conditions | Returns Policy | Intellectual Property | Cookie Policy
© 2019 – 2026 GBA Digital Media Group. All Rights Reserved
Website Design | Website Maintenance
Copyright Green Building Africa 2026.

Subscribe to our weekly Top 5 Stories
"*" indicates required fields
Δ

source