A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Mitigating peak load and heat stress under heatwaves by optimizing adjustments of fan speed and thermostat setpoint
Heatwaves are becoming more frequent and severe, intensifying cooling demand and reducing air conditioner efficiencies. This causes peaks in electricity demand that pose operational challenges to power grids. This paper provides methods to mitigate demand peaks and heat stress under heatwaves by jointly adjusting fan speeds and thermostat setpoints in buildings. The methods involve (1) learning baseline models to predict load and thermal comfort, (2) fitting perturbation models that relate fan speed and thermostat setpoint adjustments to perturbations in load and thermal comfort, and (3) optimizing peak load and thermal comfort. The methods are implementable in real buildings, providing fast, accurately predicted optimized solutions that flatten demand peaks and mitigate personal heat stress. This paper demonstrates the methodology through simulation-based case studies of a single building and a six-building neighbourhood. In case studies, the methods reduce peak load by 8–10% while maintaining occupants' thermal comfort within safe and comfortable ranges.
Highlights
This paper develops data-driven methods to reduce peak demand and mitigate heat stress during heatwaves.
The methods are designed for straightforward implementation in the field.
In case studies, the methods reduce peak demand by 8–10% while maintaining thermal comfort within safe and comfortable ranges.
To achieve the same level of peak load reduction, jointly adjusting fan speed, rather than solely thermostat setpoint, improves thermal comfort by 5% in the test case.
Mitigating peak load and heat stress under heatwaves by optimizing adjustments of fan speed and thermostat setpoint
Heatwaves are becoming more frequent and severe, intensifying cooling demand and reducing air conditioner efficiencies. This causes peaks in electricity demand that pose operational challenges to power grids. This paper provides methods to mitigate demand peaks and heat stress under heatwaves by jointly adjusting fan speeds and thermostat setpoints in buildings. The methods involve (1) learning baseline models to predict load and thermal comfort, (2) fitting perturbation models that relate fan speed and thermostat setpoint adjustments to perturbations in load and thermal comfort, and (3) optimizing peak load and thermal comfort. The methods are implementable in real buildings, providing fast, accurately predicted optimized solutions that flatten demand peaks and mitigate personal heat stress. This paper demonstrates the methodology through simulation-based case studies of a single building and a six-building neighbourhood. In case studies, the methods reduce peak load by 8–10% while maintaining occupants' thermal comfort within safe and comfortable ranges.
Highlights
This paper develops data-driven methods to reduce peak demand and mitigate heat stress during heatwaves.
The methods are designed for straightforward implementation in the field.
In case studies, the methods reduce peak demand by 8–10% while maintaining thermal comfort within safe and comfortable ranges.
To achieve the same level of peak load reduction, jointly adjusting fan speed, rather than solely thermostat setpoint, improves thermal comfort by 5% in the test case.
Mitigating peak load and heat stress under heatwaves by optimizing adjustments of fan speed and thermostat setpoint
Zhang, Zhujing (author) / Kircher, Kevin J. (author) / Cai, Yuan (author) / Brearley, Jonathon G. (author) / Birge, David P. (author) / Norford, Leslie K. (author)
Journal of Building Performance Simulation ; 16 ; 493-506
2023-07-04
14 pages
Article (Journal)
Electronic Resource
Unknown
A reinforcement learning approach for thermostat setpoint preference learning
| Springer Verlag | 2024
| British Library Conference Proceedings | 1997
| British Library Online Contents | 1997