- Title
- Sequential decoupling for managing DERs in decentralized energy systems
- Creator
- Ahmad, Ashfaq
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2020
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Environmental concerns, technological advancements, regulatory changes, emerging business models, and sustainability and resilience issues have brought distributed energy resources (DERs) into the core of future deployment of an energy infrastructure. In a broad sense, the DERs encompass a set of technologies and systems operating closer to grid customers such as renewable energy resources, controllable non-renewable energy resources, energy storage devices, flexible loads including electric vehicles, and software-based control systems. Rapidly increasing investment in new DER deployment indicates a major shift from the centralized electrical grid to a decentralized energy network, unlocking a bottom-up energy management approach. However, the growth in DER deployment brings new management and control challenges to the power grid such as intermittent generation of renewables, unpredictability of customer load, overhead costs of controllable generators, and degradation cost of energy storage devices. Energy management system (EMS) has been recognized as a potential tool to monitor, optimize and control the DERs at different customer premises such as electric vehicles, smart homes and smart buildings, while tackling the challenges associated with the penetration of DERs into the power grid. This study initially examined the existing EMS-based models to determine their effectiveness for real-time management of DERs at different end-user premises. Here, three potential application areas of end-user premise are considered namely electric vehicles, smart homes and smart buildings. The main objective is to investigate the complex interactions between different components of the system, so that a next generation EMS-based model can be designed for each application area that is useful for realistic customer energy needs, applicable in general scenarios, and suitable for practical implementation. Attention has also been paid to the system architecture in the three considered application areas. Irrespective of the application area, this study found that the existing EMS-based models are not suitable for real-time joint management of DERs due to lack of next generation components in system architecture, long-term modelling approach, requirement of a-priori knowledge on system inputs, and coupling of control actions with each other and over time. The lack of next generation components in system architecture limit the capability of the existing models to meet the evolving energy and comfort needs of the future customers. The long-term modelling approach may not suitable for real-time energy management as customers may prefer a specific period of time during which a cost saving mechanism is defined with respect to their current energy and comfort needs. The a-priori knowledge on system inputs may not be available in general or real-time scenarios because the system inputs may all fluctuate in a random and arbitrary manner with their statistics likely being non-stationary which are very difficult to predict accurately. The coupling of control actions makes it highly challenging to obtain optimal energy management decisions suitable for practical implementation. In this thesis, next generation system architectures (NGSAs) based real-time EMS models are proposed for managing DERs at the three considered customer premises. First, an NGSA based EMS model is presented for managing electrical DERs in electric vehicles. The NGSA for electric vehicles consists of solar integrated electric vehicles with an on-board energy storage battery that can procure energy from different charging stations at home, parking lot and other public locations. The charging stations are also equipped with renewable energy sources. Here, the goal is to minimize an average aggregated system cost through a real-time joint optimization of electric vehicle’s energy procurement price, load scheduling delays, photovoltaic sufficiency in terms of locally generated renewable energy mix, and battery degradation. Second, an NGSA based EMS model is presented for managing electrical DERs in micro-grid connected smart homes. The NGSA for smart homes consists of a stand-alone micro-grid in which each home is powered by a roof-top photovoltaic generator and is connected to a common peaker generator. Each home is also connected to a common energy storage battery system to store energy from the power sources and supply it to the households. The goal is to minimize an average aggregated system cost through a real-time joint optimization of energy storage battery management, load scheduling, and energy procurement process from the peaker generator. Here, the concept of block duration is introduced to optimize the energy procurement cost from a peaker generator. Third, an NGSA based EMS model is presented for managing both electrical and thermal DERs in grid-connected smart buildings. The NGSA for smart buildings consists of electrical and thermal energy sources, storage systems, and loads. Where, the sources include a utility grid, a building integration photovoltaic source, and controllable generators (i.e., a combined heat and power system, and a gas boiler); the storages include both electrical and thermal energy storage systems; and the loads include both electrical and thermal loads. The goal for managing both electrical and thermal DERs in a grid connected smart building is to minimize an average aggregated system cost through a real-time joint optimization of electrical and thermal load scheduling delays, energy procurement cost from controllable generators and external grid, electrical and thermal energy storage degradation, and indoor user comfort feel (i.e., visual, thermal and air quality comfort). Fourth, the NGSA based EMS model for grid-connected smart buildings is further enhanced in three aspects: electric vehicle load is carefully integrated into the building's electrical and thermal load while addressing its load scheduling implications, a new model is added for optimal sizing of DERs in the building prior to their real-time control, and the system cost model is upgraded by taking into account environmental deterioration cost and operations-and-management cost of controllable non-renewable generators. Each of the proposed NGSA based EMS models in this thesis employs a one-slot-look-ahead virtual queue stability based Lyapunov optimization technique for a real-time management of DERs. The proposed models introduce customer oriented specific duration based modelling, rely on current states of the system inputs only, and sequentially decouple the control actions from each other and over time. As a consequence, the proposed models are useful for practical customer needs in terms energy and available budget, applicable in general scenarios especially under unknown dynamics of system inputs, and suitable for practical implementation in real-time test beds. The proposed models are validated through extensive simulations where these are tested in different real-time scenarios and under varying weather conditions. Results show the better adaptation capability of the proposed models to real-time changes in energy demand, supply, storage and comfort conditions, along with economic gains and quality-of-service assurance. Further research is recommended to improve and extend the proposed models towards large-scale integration of DERs. This may need investigations related to network stability, cyber-physical behavior awareness, software defined networking, cloud/fog computing, communication network technologies, and network security.
- Subject
- energy management; real-time optimization; sizing; control; smart grid; distributed energy resources; renewable energy
- Identifier
- http://hdl.handle.net/1959.13/1421568
- Identifier
- uon:37745
- Rights
- Copyright 2020 Ashfaq Ahmad
- Language
- eng
- Full Text
- Hits: 1923
- Visitors: 2413
- Downloads: 536
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 35 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 4 MB | Adobe Acrobat PDF | View Details Download |