Author Identifier

Nishadi Ruwandima Weerasinghe Weerasinghe Mudiyanselage: http://orcid.org/0009-0005-7646-2919

Date of Award

2025

Document Type

Thesis

Publisher

Edith Cowan University

Degree Name

Master of Engineering Science

School

School of Engineering

First Supervisor

Asma Aziz

Second Supervisor

Iftekhar Ahmad

Third Supervisor

Bassam AL-Hanahi

Abstract

The rapid growth of electricity demand, increasing energy costs, and the global aim for decarbonization have driven the need for intelligent energy management systems (EMS) in the residential sector. Multi-storey residential buildings, in particular, present unique challenges for demand side management due to their complex load profiles, common-area energy usage, and occupant behaviour diversity. Traditional Demand Side Management (DSM) programs, while successful in industrial and commercial settings, are less explored in multi-storey residential communities where user participation and load flexibility are more difficult to manage. Traditional DSM schemes in multi-storey residential settings face limitations due to the heterogeneity of occupant behaviors, lack of centralized control, and restricted access to individual load data. These factors hinder real-time responsiveness and coordinated energy optimization across shared infrastructure.

This research explores the integration of flexible load technologies, rooftop solar photovoltaics (PV), and community battery storage to enable demand response and community energy sharing within multi-storey residential buildings. A comprehensive literature review identifies high-potential flexible residential appliances and community energy-sharing models. Building on these insights, a 24-hour synthetic dataset is developed, incorporating appliance-level power consumption and occupancy-driven behavioural patterns, alongside common-area loads such as hot water systems and pool pumps, managed under strata ownership.

A building energy management system (BEMS) is developed to optimize the use of rooftop solar energy across both individual apartment and shared facility loads. Solar generation is simulated using irradiance data specific to Joondalup, Western Australia, and a shared battery energy storage system is employed to further enhance local energy consumption. Battery discharge is proportionally distributed among apartments based on their respective hourly load profiles.

Although several metaheuristic and open-source solvers are available, the energy scheduling problem in this study employs a MILP formulation to ensure exact, globally optimal solutions. The Gurobi Optimizer was selected for its superior solving capacity: it consistently achieves the fastest solve times on large LP/MIP problems Gurobi Optimization, handles complex models over multi‑core systems, and incorporates advanced presolve, cut generation, and tuning capabilities . Moreover, Gurobi offers robust numerical reliability, extensive Python API support, and academic licensing all of which facilitate efficient and scalable optimization for the Building EMS. The optimization framework is designed to minimize total electricity costs and grid reliance while maximizing solar self-consumption. It incorporates time-of-use electricity tariffs, flexible appliance scheduling, coordinated solar-battery operation, and community energy sharing with equal battery capacity allocation. Electric vehicle (EV) charging is treated as a flexible communal load, and ownership-aware constraints ensure fair access to shared energy resources among all residents.

Simulation results demonstrate that the proposed EMS significantly reduces peak grid demand and increases the utilization of locally generated renewable energy. The framework offers a scalable, ownership-aware solution that supports equitable energy resource distribution, cost efficiency, and enhanced resilience advancing the goal of sustainable, low-carbon urban living.

Comments

Author also known as Nishadi Weerasinghe

DOI

10.25958/swry-q180

Access Note

Access to this thesis is embargoed until 2nd September 2027

Available for download on Thursday, September 02, 2027

Share

 
COinS