Modeling and control optimization of aggregated EERs for the implementation of smart grid functions, peak load shifting.
Loading...
Date
2025-10
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of New Brunswick
Abstract
Demand Response (DR) represents a transformative paradigm in modern smart grids, leveraging the energy storage capabilities of distributed small-rating devices and bidirectional hybrid resources. This framework establishes a mutually beneficial agreement between utilities and end-users. Utilities gain the ability to manage rapid increases in electrical demand, mitigate peak loads, and enhance grid stability, while end-users are incentivized through reduced energy costs, improved power quality, and maintained comfort levels.
Thermostatically controlled loads (TCLs) are particularly well-suited for DR implementation due to their widespread deployment, significant energy storage capacity, and substantial energy consumption. Centralized control is facilitated through aggregators, which utilize advanced communication networks and remote-control units to optimize activation periods and capacity sharing.
Aggregator systems rely on accurate modeling, robust forecasting, and reliable communication networks. However, technical and systematic challenges in management necessitate further research. This thesis focuses on three primary objectives related to peak shaving services:
1. maximizing aggregator-shared capacity while addressing technical complexities in load control.
2. Analyzing the impact of DR actions on distribution networks and quantifying demand response mismatches.
3. Providing comparative studies to assess the impact of forecasting errors and uncertainties on DR programs.
The proposed DR system, characterised by a hierarchical structure comprising virtual power plants (VPPs), aggregator, and end-user levels, is validated through extensive simulations and operational scenarios. Evaluation criteria include comprehensive analysis, impact indices, performance metrics, and comparative calculations to assess system performance and effectiveness in peak shaving.
Results demonstrate that the proposed DR system and control algorithms significantly enhance capacity provision compared to conventional algorithms. However, continuous aggregator device control affects customer comfort and induces grid voltage fluctuations at the distribution level. Furthermore, while the DR program exhibits robustness against uncertainties in certain parameters, other can impede its effective implementation.