Optimal design and scheduling for Behind-the-meter PV-battery energy storage systems
Loading...
Repository DOI
Publication Date
2024-09-11
Type
doctoral thesis
Downloads
Citation
Abstract
The ongoing pursuit of a green and sustainable future has led researchers to explore new areas of energy engineering, particularly for power system end-users. In this context, an increasingly attractive topic is the achievement of net-zero through behind-the-meter (BTM) energy generation and storage. One viable BTM option is the installation of rooftop photovoltaic (PV) panels coupled with battery energy storage systems (BESSs). An optimally designed and controlled BTM PV-BESS can provide effective services to both end-users and utilities. Otherwise, it can impose unnecessary expenses on their owners. This thesis contributes to the development of optimization models for PV-BESSs, encompassing long-term system design and sizing, and short-term and real-time scheduling strategies, aiming to minimize payback period, bill costs and BESS degradation. To achieve this, innovative models are developed using a range of optimization techniques, including linear and mixed-integer linear programming, heuristic modeling, scenario-based stochastic programming, and Machine Learning (ML)-based modeling. To account for uncertainties in the problem’s input data, stochastic scenario generation models are proposed using statistical and ML techniques, utilizing information from real-world projects. The motivations, design constraints and methodologies of optimization models are presented and discussed in detail, and results from simulations under different scenarios are thoroughly investigated. The results from different chapters of this thesis hold the potential to benefit government agencies, utilities, and policymakers in designing incentivizing schemes that encourage end-users to purchase and install BTM resources on their premises. The developed models can be applied to PV-BESS systems of any scale. However, in this thesis, all the models are tested on residential units because data was only available from residential houses during this PhD research.
Publisher
University of Galway
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International