Please use this identifier to cite or link to this item: http://hdl.handle.net/10266/6687
Title: Design of a DC Net Meter in Vehicle to Grid Technology
Authors: Yadav, Kratika
Supervisor: Singh, Mukesh
Keywords: Electric Vehicle;Vehicle to Grid;DC Net meter;Time of Use Tariff
Issue Date: 23-Jan-2024
Abstract: The growing acceptance of electric vehicles as a sustainable transportation solution has led to higher demand for efficient and reliable charging infrastructure. However, multiple chal- lenges need to be addressed before the seamless integration of EVs with the existing grid can be achieved. Firstly, the current grid infrastructure may not be fully prepared to meet the in- creasing electricity demand resulting from widespread EV adoption. Charging multiple electric vehicles at once, particularly during peak hours, can strain the system and result in power out- ages or voltage instability. Another important consideration revolves around the requirement for smart metering infrastructure capable of monitoring the bidirectional flow of electricity be- tween the grid and the vehicles. Further, this infrastructure plays a pivotal role in enabling vehicle-to-grid technology to provide a more accurate and comprehensive overview of elec- tricity consumption. Furthermore, the conventional centralized charging infrastructure raises concerns regarding scalability, availability, and accessibility. In response to these challenges, four distinct methodologies and schemes have been proposed. In the first approach, a bidirectional DC net metering system has been proposed for V2G technology. The study focusses on addressing the challenges associated with AC-side metering in EV charging system. The objective is to provide precise measurements for end customers by placing the DC net meter on the battery side. Further, the proposed metering system is designed to comply with international standards and incorporates bidirectional power transfer, real-time data communication, and a net metering scheme for accurate cost calculation. Additionally, the research emphasises the potential for integrating dynamic pricing structure and utilizing the Internet of Things (IoT) to manage data from multiple DC net meters. The outcomes of this study contribute to the development of a smart DC net meter for V2G that enhances billing accuracy for EV charging operations and creating opportunities for future enhancements. The second study investigates the role of dynamic pricing in optimizing power grid opera- tions in the face of increasing EV adoption. It introduces a time-of-use (TOU) pricing frame- work that incorporates critical peak pricing (CPP) and peak time rebate (PTR) to coordinate iv EV charging and reduce electricity expenses. The research underline the significance of fac- tors such as peak/off-peak hours, state of charge (SOC), dynamic pricing, and the presence of V2G-enabled stations in achieving efficient charging and discharging. The proposed approach demonstrates better performance compared to previous models in terms of benefits, cost sav- ings, and accuracy. Furthermore, the study seeks to emphasize the feasibility and advantages of the TOU-CPP/PTR tariff structure and the important role of dynamic pricing in managing EV charging. Moreover, in the third study, a decentralized charging scheduling approach is put forth as a solution to address the challenges arising from the rapid expansion of electric vehicles. The proposed approach optimizes the charging schedules in a decentralized manner with the goal of minimizing the electricity costs. In addition, the findings marked the influence of EV charging on overall consumption of electricity and highlights the importance of understanding the dy- namics between decentralized charging and the base load curve. Additionally, it is highlighted that ongoing research and development in charging algorithms are essential to seamlessly inte- grate EVs into the grid and explore advanced technologies for enhancing grid flexibility. Lastly, an attention-based deep learning model for load forecasting has been introduced to address the challenges posed by the growing prevalence of EVs. The proposed model of- fers utilities the capability to dynamically adjust energy supply in real-time, thereby preventing system overloads or underutilization. With the increasing adoption of EVs, power grids face greater difficulty in maintaining a delicate balance between supply and demand. Addition- ally, the fluctuations in demand resulting from EV charging can strain the grid’s infrastructure. The study showcases the substantial improvements achieved in energy management for public EV charging infrastructure through efficient data preprocessing and the utilization of cutting- edge deep learning algorithms such as LSTM and GRU. The demonstrated accuracy and effec- tiveness of these models open doors for further exploration in energy management, predictive maintenance systems, and real-world testing, collectively addressing the evolving landscape of electric vehicle integration and grid stability.
URI: http://hdl.handle.net/10266/6687
Appears in Collections:Doctoral Theses@EIED

Files in This Item:
File Description SizeFormat 
FinalThesis_KY.pdfPhD Thesis18.01 MBAdobe PDFView/Open    Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.