Energy management strategies for smart grids with renewables

Theoretical thesis.Bibliography: pages 193-210.Chapter 1 Introduction -- Chapter 2 Literature Review -- Chapter 3 Performance Analysis of an Experimental Smart Building -- Chapter 4 Energy Trading in Local Electricity Market with Renewables -- Chapter 5 Optimal Price Based Control of HVAC Systems -- Chapter 6 Conclusion and Future Work.With increasing environmental concerns raised from fossil fuel sources, the prominent feature of next - generation smart grids is to supply power from clean/renewable energy sources (i.e., solar, wind and fuel cell, etc) in order to provide economic, environmental, reliability and security benefits. To achieve these goals , future smart grids will work in highly complex and dynamic environments and will have small - capacity distributed renewable energy generators (DREGs) with non - dispatchable and intermittent characteristics. Moreover, the utilization of DREGs on a large - scale helps to flatten peak demand to avoid substantial overcapacity in the size of a power system due to high aggregated peak demand. However, DREGs need to manage, and they required interaction with each other, with storage systems and an energy provider for improved asset utilization and energy efficiency. In this context, an efficient demand-side management system (DSMS) is essential for coordinate control of DREGs and responsive loads to maximize the system's utilization and reliability in a smart grid. Fundamentally, demand-side management (DSM) is a process of shifting/reshaping electrical loads and utilizing new technologies to reduce power bills, overall operational costs and increase energy efficiency. This thesis addresses the challenges of developing a framework for optimal DSMS by modeling the energy usage behavior of self-interested distributed entities through studying the propriety DREGs and consumers in a smart grid. The major contributions of this research are given below. The first contribution of this research is to develop an algorithm for analyzing the performance of an experimental smart building through real-time data analysis, and then recommend possible measures to improve its energy efficiency. It focusses on the performance gap in terms of energy efficiency and the criticalities related to the characteristics of chosen devices and demand management strategies adopted. In addition, new technologies (to enhance DREGs production), coordinated measures (to improve building energy management system) and transactive control (to control the building's responsive load) are proposed. The scientific analysis of proposed recommendations for an intelligent energy management system demonstrates significant energy and cost savings for smart buildings. The second contribution of this research is to present a three-level hierarchical energy-trading framework for encouraging the owners of DREGs to voluntarily take part in an energy trading process. The developed strategy captures the complex interactions between the owners of geographically DREGs and the aggregator in the smart grid using a non-cooperative contract theoretic approach. Moreover, a dynamic pricing scheme is developed that the aggregator can utilize to incentivize the owners of DREGs and a distributed algorithm is proposed to enable the energy-trading process. Various categories, types, and constraints of DREGs, different trading scenarios and wholesale price impact on trading are considered in the analysis for practical applications. The solution of the developed scheme shows that socially optimal energy management for both trading partners can be achieved. The third contribution of this research is to develop an occupant's comfort aware energy imbalance management scheme for efficiently curtailing responsive loads of commercial buildings with the market price. The aim is to reduce the aggregated and peak demand to deal with energy imbalance problem in case of DREGs intermittency and/or power shortage from the grid while providing the desired quality of service. To achieve this goal, an intelligent and new price-based demand response (PBDR) control strategy is proposed to optimize the responsive load scheduling. Occupants' varying thermal preferences in the response of price signals are considered and modeled using the artificial neural network (ANN) to integrate into the optimal scheduling problem. The performance of the proposed management techniques is tested in real Australian power distribution networks under real load, weather conditions, and electricity tariff structure. The developed models, algorithms, and techniques can capture the different cost-benefit trade-offs that exist for efficiently managing buildings energy in a smart grid. These strategies have shown significant performance improvement when compared with existing solutions. The work in this thesis demonstrates that modeling power usage behavior of distributed entities in a smart grid for robust DSMS is both possible and beneficial for increasing the energy efficiency of smart buildings in a smart grid.1 online resource (xxv, 211 pages) colour illustration