.png){kind=icon}
.png){kind=icon}
Return
Data-oriented distributed overall optimization for large-scale HVAC systems with dynamic supply capability and distributed demand response
.png){kind=icon}
PRE
AI
DOI：10.1016/j.buildenv.2022.109322.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Operational optimization of heating, ventilation, and air-conditioning (HVAC) systems is crucial to achieving economic and environmental benefits. However, owing to the dynamic supply capability (DSC) and distributed demand response (DDR), large-scale HVAC systems that integrate multiple distributed sources with demanders are increasingly being developed to resemble distributed energy systems (DESs); thus, the traditional centralized strategies are now facing unprecedented challenges. With the application of wireless communication and sensor network technology, distributed schemes that allow the modular integration of an arbitrary number of suppliers and demanders have attached widespread interest of researchers. However, existing distributed optimization schemes that focus on either multi-zone temperature or equipment group efficiency are unilateral and non-interfering. In this work, we developed a data-oriented distributed overall optimization strategy for large-scale HVAC systems with DSC and DDR. We first suggested an agent-based distributed overall optimization frame-work that defines the roles of participants in distributed optimization and clarifies their duties, mechanisms, and relationships with each other, thus bridging the supply-side and demand-side decisions. Then, we exploited state-of-the-art data-driven models and designed a bi-level optimization algorithm, where the heuristic algorithm works as the lower-level scheme and the Nash-optimization algorithm acts as the upper-level scheme, to implement the distributed overall optimization. Last, a hardware-in-the-loop simulation was performed, and time-series data were collected to validate our proposed optimization strategy. Compared with centralized strategies, it can obtain better operational performance by paying a lower computational cost. This work pro -vides a reference for optimizing increasingly complex HVAC systems.
Keywords:
Distributed optimization strategy Differential supply capacities Distributed demand responses Bi-level optimization algorithm Supplied pressure head reset
Differential supply capacities
Distributed demand responses
Bi-level optimization algorithm
Supplied pressure head reset
Journal
IF:
7.6
Papers:
1.3W
Citations:
6.6W