The proliferation of demand-side management and demand response programs can be attributed to electrification and the rise in the utilization of renewable energy sources. The adoption of dynamic pricing structures has necessitated the utilization of advanced techniques for dynamic simulation and optimization of building operation. This is aimed at reducing economic costs, optimizing renewable energy utilization, and facilitating collaboration among multiple households in virtual clusters known as virtual energy communities. Within this framework, energy aggregators coordinate the individual houses and deliver an equivalent aggregated power demand to grid operators. However, the scalability of these control-oriented methods is still a barrier to their widespread use because of the large number of parameters to identify. The present study explores a model order reduction application that employs a partitional clustering technique, specifically k-Means. This approach identifies equivalent aggregated thermal zones that are formed by combining zones with similar temperature patterns and setpoint profiles. Building models are identified for the aggregated thermal zones using an inverse gray box modeling technique, and the control routine is implemented within a model predictive control (MPC) framework to determine the optimal setpoint profiles based on a day-ahead weather forecast. This approach involves deploying fewer control variables (setpoint profiles) than the actual number of rooms using the aggregated zones, facilitating the optimized operation. The performance of the proposed model is evaluated by comparing it to reference models, in which model reduction is implemented by treating each floor as a separate thermal zone. The current investigation uses a detached house in Trois-Rivieres, Quebec, to demonstrate the proposed methodology's feasibility. The chosen case study uses detailed measurements for every thermal zone and the architectural plans of the building. The features mentioned above are employed during the model validation section to justify the underlying assumptions and provide commentary on the obtained results.
| File Size : | 1
file
, 3.3 MB |
| Note : | This product is unavailable in Russia, Belarus |
| Number of Pages : | 10 |
| Product Code(s) : | D-CH-24-C069 |
| Published : | 2024 |
