Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/66236
Title: Energy storage management for renewable energy systems
Authors: Feng, Xue
Keywords: DRNTU::Engineering
Issue Date: 2016
Source: Feng, X. (2016). Energy storage management for renewable energy systems. Doctoral thesis, Nanyang Technological University, Singapore.
Abstract: Excess dependence on fossil fuels and hence large amount of carbon emission overloads the atmosphere of the earth and causes global warming which poses negative impacts on the sustainability development of the human being. The increasing carbon emission is largely due to the deployment of the carbon intensive electricity generation based on fossil fuels. The replacement of fossil fuel based generators by renewable energy generators is able to significantly reduce the carbon emission, mitigate the global warming and bring benefits to the public health. The renewable energy supply which is vastly available and inexhaustible in nature, has a great potential to take up a greater share of the energy supply from a technical perspective. Still, there exist key issues to be addressed. The renewable generation is normally highly reliant on the weather conditions, such as the photovoltaic generation is directly related to the solar irradiation and the wind generation to the wind speed. The unpredictability, inconsistency and unreliability of the renewable energy generation, with the absence of auxiliary mitigation measures, make them less attractive for large-scale implementations. The economic viability of the renewable energy integration is another concern. The renewable energy generation is much smaller in quantities as compared with traditional generators, suggesting that a larger number of renewable energy facilities need to be put in place to reach the same amount of the electricity generation. This will incur a high investment cost. As such, it is crucial to optimize the utilization of the renewable energy so that the high capital expenditure can be compensated. Energy storage systems, which can be used to store the surplus generation from renewable sources and later deliver the energy when generation is insufficient due to weather conditions, provide effective solutions to the intermittent issue and economical concerns with renewable energy generation. The supply-demand mismatches can be dynamically compensated by the energy storage systems under appropriate energy management strategies, in order to ensure the reliability of the renewable energy supply systems. In this thesis, the research scope is devoted to optimally apply the energy storage system in mitigating the reliability problems of the renewable energy generation so that a continuous and reliable renewable energy supply is accomplished, taking into account of the advantages and constraints of the energy storage technologies. PV generation, which is one of the most rapidly expanding and commercially exploited renewable energy source, is used as an example to be considered. The research methods and findings are transferable to other renewable sources based on the common feature of intermittency. Meanwhile, battery energy storage systems and in particular lithium ion batteries are the focus of this research due to the high efficiency, high energy density and broad applicability. The first step to optimize the battery energy storage usage is to thoroughly analyze the characteristics associated with the batteries of interest. Based on the analysis, a sophisticated battery model, describing the external dynamic behaviors originated from complex internal electrochemical reactions, needs to be formulated. The model to be constructed should hold such features as a high accuracy under dynamic conditions considering the intermittency introduced by the renewable energy generation and the changing load consumption, a marginal computational burden to fit the system-level simulations where compatibility with other electric components is required, and a precise thermal description to relieve the safety concerns. By reviewing the relate studies, an abstract dynamic model based on a simple equivalent circuit is proposed for a lithium ion battery. Additional elements are introduced in order to reflect the complicated features of the battery under varying currents, state of discharge changes and different temperature conditions. The temperature rise caused by excess internal heating, which may degrade the battery efficiency and incur safety concerns, is particularly tracked by incorporating a sophisticated temperature prediction approach. The heating sources are comprehensively included to ensure a reliable temperature prediction. The model performance in term of describing the battery charge/discharge behaviours as well as the temperature rise prediction are validated by the experimental tests carried out on the \textit{Ultralife} UBBL 10 lithium ion battery. It's noted that the calculations involved are conveniently finished in a short period of time. The dynamic, effective and simple battery model with the accurate temperature prediction, which is suitable for system level simulations, paves the way for investigating and optimizing the utilization of batteries in the renewable energy systems. The fluctuating weather conditions result in frequent presence of highly volatile power components in the renewable energy systems. The power variations need to be handled by the battery energy storage to ensure the system reliability. However, the battery efficiency and lifetime ar significantly decreased by the alternating charge/discharge alterations and sudden power surges. In addition, the safety concerns on the excess heating of the battery may be caused by the abuse of battery. In order to make full use of the strength of the battery which is the high energy density while avoiding the exposure of batteries to high power variations, another storage device with the complimentary features can be combined with the battery to form a hybrid energy storage (HES) which obtains a high energy density and a high power capability at the same time. Ultracapacitors, with a high power capability, an ultra-fast response speed and a long cycle life are adopted and combined with the battery. The HES can exploit the strengths from two storage devices while offset the weakness with individual devices. In this thesis, the HES is used to maintain the system stability by compensating the imbalance between the PV generation and the load consumption. The device models including the dynamic battery model proposed are incorporated in the system simulations. A fine-grained intelligent control strategy with multiple operation modes is designed to distribute the power demand between the batteries and the ultracapacitors considering the real time power mismatches. Based on the PV irradiation data collected from the field and the Singapore-pattern load data, the robustness of the HES under the intelligent control strategy is validated in both short-term and long-term scenarios. The HES with the proposed control strategies demonstrate flexibility to cope with transient disturbances and reliability under long-period operations. Compared with two conventional methods, the proposed control method for the HES demonstrates improved efficiency of involving the ultracapacitors to handle power spikes while protecting batteries from misuse. The economic issues on high investment cost and replacement cost of batteries need to be solved for a broader adoption of the battery energy storage in renewable energy applications. An effective energy management system considering the capacity fade of the battery is pressingly required. To begin with, an accurate and dynamic lifetime model of the battery which can track the battery capacity fade needs to be formulated in the first place. Based on the existing research work, a practical capacity fade model which aims at tracking the real time capacity loss is proposed and validated by intensive cycling tests. The real time capacity fade model is incorporated with the electrical circuit model to give a more thorough description of the battery. The two models mutually exchange information and update parameters. An online energy management strategy considering the battery capacity fade is then proposed for a battery bank which is used to compensate the supply-demand mismatches. Individual differences among the batteries constituting the battery bank are fully considered in the optimization process to enhance the economic effectiveness of the system. Evaluation results on various weather conditions and different battery statuses exhibit the flexibility and viability of the online energy management strategy.
URI: http://hdl.handle.net/10356/66236
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Theses

Files in This Item:
File Description SizeFormat 
Thesis_Mar18.pdf
  Restricted Access
5.43 MBAdobe PDFView/Open

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.