Novel real-time data management modules for general community-level grid based on centralized V2G System
Year: 2026
Type: dissertation
Abstract: Vehicle-to-Grid (V2G), as an emerging form of temporary energy storage, has been widely studied for its potential applications in grid management, aiming to support peak shaving and power balancing while mitigating the impact of the variable and intermittent output of renewable energy on irregular grid load distributions. Previous research on V2G management and scheduling has seen extensive development at the microgrid level. However, the complexity of electric vehicle (EV) usage behaviours poses significant challenges, limiting the direct applicability of these microgrid-based V2G strategies to community- and region-level centralized frameworks. To bridge this gap, a centralized V2G scheduling framework was developed and implemented in Python as a simulation-based platform. The framework was designed with the overarching objective of achieving extreme load balancing within community-scale zero-carbon grids. To ensure realism and applicability, the underlying grid model was constructed using operational data from the Nottingham Trent Basin community energy project, a flagship zero-carbon initiative in the UK, in combination with empirical driving and charging behaviour statistics drawn from the MyElectricAvenue project, a national study of electric vehicle users. Three distinct scheduling strategies were evaluated to investigate the operational potential of centralized V2G systems under varying assumptions of vehicle availability and user behaviour. The first strategy prioritized the utilization of active EVs—defined as vehicles that exhibit either driving or charging behaviour within a given day—for meeting load balancing requirements, with idle EVs serving as supplementary resources when the active fleet alone was insufficient. The second strategy preserved the charging behaviour of active EVs entirely and relied exclusively on idle vehicles for V2G operations, a configuration that to some extent reflects the technical feasibility of simplified battery-swapping station paradigms. The third strategy treated all EVs as fully dispatchable resources, effectively emulating the behaviour of a large-scale stationary storage system and representing an extreme boundary condition for centralized scheduling. This latter scenario further allowed for an exploration of the substitutability of EV-based flexibility against dedicated battery storage facilities. Results indicate that under current community grid conditions and typical EV usage parameters—including 3.3 kW bidirectional power transfer, 24 kWh battery capacity, full EV ownership, and a 54.87% idle rate—the proposed framework effectively achieves community-level load balancing. Economically and environmentally, Scenarios I and II demonstrate a 49.78% reduction in annual electricity costs and a net negative carbon footprint, confirming the robustness of the control logic. Further sensitivity analysis identified parameter configurations that enhance the efficiency of V2G scheduling. Specifically, when higher charging and discharging power levels are combined with larger EV battery capacities, the number of vehicles required to achieve extreme load balancing within the grid can be significantly reduced, thereby improving the effectiveness of each EV in the scheduling process. However, these gains in operational efficiency are accompanied by higher economic costs and a modest decline in carbon performance, highlighting the trade-offs inherent in configuring system parameters for community-scale V2G applications. Collectively, these findings demonstrate that systematically integrating idle EVs into centralized V2G frameworks provides a scalable, practical approach for balancing community-level power loads, minimizing costs, and advancing carbon neutrality, offering a robust, data-driven foundation for future regional-scale V2G deployment and smart grid development.
Author: Ru, Jie
Institution: University of Nottingham
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7218300829
