Using AI to optimise energy flow management in urban electric vehicle charging

在这个项目中，Lesla Ltd和克兰菲尔德大学将开展一项可行性研究，以确定使用人工智能解决方案来优化形成能源微电网的众多Lesla街头电动汽车充电器的能量流管理的优点和可行性。城市电动汽车充电对于实现英国Net-Zero立法目标至关重要。超过80%的电动汽车在家中充电，但英国38%的家庭无法使用非街道停车，他们将依赖公共街道充电，而目前公共街道充电服务严重不足-现在只有不到2%的电动汽车依赖于街道停车。Lesla路缘充电系统解决了这个问题。它鼓励电动汽车在夜间停放时一直充电，利用最优的能源定价以及先进电网服务的收入，使能源更实惠，这最近对许多英国家庭来说尤其重要。许多城市充电器的优化能量流管理涉及许多能源市场参与者之间的多方面关系，每个参与者都有自己的、有时相互矛盾的需求，这种需求并不总是能够可靠地预测出来。这个项目将研究使用人工智能来预测和优化这些由电动汽车和多居民城市建筑组成的微电网内的能量流动的可行性：*确保电动汽车可以用最低价格的能源充电，鼓励在可再生能源充足的时候充电，以及*将电动汽车的充电从电网需求高峰期转移，以确保在不增加整体邻区电网连接容量要求的情况下可以使用现有的电网连接。*使社区能够从先进的电网服务和负载平衡工具中获益。人工智能解决方案应该能够提供一个自适应的能量流管理系统，它将不断学习建筑物居民和电动汽车司机的行为，以适合在新的社区部署，并增加新的充电设备，以便在未来大规模部署。可行性研究将：*为城市微电网内支持人工智能的Lesla电动汽车路缘充电器管理提供技术、经济、财务、法律和环境考虑因素的评估。*与利益相关者接触，-证实用户需求及其对能源使用激励的响应-根据经验数据建立可预测的充电配置文件。*在进行后续更大的项目之前确定成功的可能性。

In this project Lesla Ltd and Cranfield University will carry out a feasibility study to determine the merits and viability of using Artificial Intelligence solutions to optimise energy flow management of numerous Lesla on-street electric vehicle chargers forming energy microgrids.Urban EV charging is essential to reach UK Net-Zero legislative targets. More than 80% of electric vehicles charge at home, but 38% of households in UK don't have access to off-street parking, and they would depend on public on-street charging, which currently is critically underserved - less than 2% of EVs rely on on-street parking now.Lesla kerb charger system solves this problem. It encourages electric cars to be plugged in all the time while parked overnight, taking advantage of best energy pricing as well as income from advanced grid services, making energy more affordable, which lately has been particularly relevant for many households in UK.Optimising energy flow management for many urban chargers involves multifaceted relationships between many energy market players, each with their own, sometimes contradictory requirements, which not always can be reliably predicted. This project will research the feasibility of using Artificial Intelligence to forecast and optimise the energy flows within these microgrids formed by electric vehicles and multi-dweller urban buildings:* to ensure that the electric cars can be charged with the lowest-priced energy, incentivising to charge when renewable energy is abundant, and* shifting the EV charging away from periods of high grid demand, to ensure that existing grid connections can be used without increasing the overall neighbourhood grid connection capacity requirements.* to enable community to earn from advanced grid services and load balancing tools.Artificial intelligence solution should be capable of providing a self-adapting energy flow management system, which would continually learn from the behaviours of buildings' residents and electric vehicle drivers, to be suitable for deployment in new neighbourhoods, and adding new charger devices for mass deployment in the future.The feasibility study will:* provide an assessment of technical, economic, financial, legal, and environmental considerations for AI enabled Lesla EV kerb charger management within urban microgrid.* engage with stakeholders, - to substantiate user needs and their responsiveness to energy use incentivisation, - to establish predictable charging profiles from empirical data.* determine the likelihood of success before conducting a subsequent larger project.