Author Identifier
Pema Dorji: http://orcid.org/0009-0003-3243-3337
Date of Award
2026
Document Type
Thesis - ECU Access Only
Publisher
Edith Cowan University
Degree Name
Master of Engineering Science
School
School of Engineering
First Supervisor
Stefan Lachowicz
Second Supervisor
Octavian Bass
Abstract
The rapid integration of distributed generation into modern distribution systems necessitates advanced methodologies for optimal placement, sizing, and control to ensure reliable, efficient, and sustainable operation. This thesis addresses the challenges associated with integrating photovoltaic distributed generation and battery energy storage systems into unbalanced distribution networks, where conventional planning approaches often rely on balanced-system assumptions and overlook the impact of reactive power support and harmonic distortions. The research proposes a comprehensive multi-objective optimization and control framework to enhance technical, economic, and environmental performance, validated through rigorous simulation on IEEE 13- and 37-bus unbalanced test feeders. A core contribution of this work is the development of an integrated optimization framework that simultaneously minimizes active power losses, total costs, and carbon emissions while considering both active and reactive power injection strategies. By systematically evaluating P-only and PQ configurations, the study demonstrates that reactive power support significantly improves voltage stability, reduces localized losses, and mitigates carbon emissions, underscoring the critical role of reactive power in unbalanced systems.
The framework employs the Reptile Search Algorithm as the primary optimization technique, with Particle Swarm Optimization serving as a benchmark. Comparative analyses reveal that Reptile Search Algorithm outperforms Particle Swarm Optimization in convergence speed, robustness, and solution quality, particularly under high-dimensional, unbalanced conditions. To address harmonic distortions arising from inverter-based DG integration, a stationary-frame Proportional Resonant (PR) controller is implemented and optimized using hybrid meta heuristic approaches, including Particle Swarm Optimization, Genetic Algorithm, and a hybrid Particle Swarm-Genetic Algorithm approach. The optimized PR controller achieves substantial harmonic mitigation, maintaining total harmonic distortion below 1.1% while ensuring precise current tracking under unbalanced load conditions. This dual focus on optimal DG deployment and advanced control design establishes a unified framework for both planning and operational reliability in realistic distribution networks.
The research emphasizes the importance of unbalanced network modeling, demonstrating that reliance on balanced assumptions can yield misleading results. Sensitivity and scenario analyses confirm the robustness of the proposed methodologies under varying load profiles, DG penetration levels, and renewable generation uncertainties. Limitations include the focus on IEEE test feeders, simplified battery system modeling with constant power output, and exclusive use of stationary-frame PR controllers. These limitations suggest future research directions, such as extension to larger, heterogeneous networks, incorporation of detailed battery models, exploration of adaptive or predictive control strategies, and integration of AI-driven optimization techniques for improved scalability and intelligence. Overall, this thesis establishes a technically rigorous, multi-objective methodology for integrating distributed generation into unbalanced distribution systems, providing actionable insights for enhancing voltage stability, reducing losses and emissions, and mitigating harmonics, thereby supporting sustainable, high-performance network operation.
Access Note
Access to this thesis is embargoed until 13th September 2027
Recommended Citation
Dorji, P. (2026). Multi-objective optimization of size and siting of PVDG/BESS to enhance the stability and power quality of distributed generation integrated into unbalanced distribution system. Edith Cowan University. https://doi.org/10.25958/npgz-gp82