Impact studies of electric vehicles and behind-the-meter photovoltaic generation in electric power distribution networks based on artificial intelligence

电动汽车（EV）和光伏（PV）发电在家庭中的采用在短期内迅速而大规模地增加。 因此，公用事业公司面临着有效管理这些资源的挑战，以帮助减轻其连接到电网时运行的不良副作用。 解决这些问题的现有方法依赖于对电动汽车和PV（作为分布式能源或DER的一部分）行为的准确预测，对客户和电网基础设施的详细了解，所有这些都使其有效部署变得复杂。 因此，该项目的主要目标是开发基于数据驱动的人工智能（AI）工具，消除对配电网络某些部分的详细建模的需求，并解决准确预测的问题。 这些工具将被开发用于控制配电网中的DER，将允许研究DER的影响并协调其操作以在公用事业网络上提供需求响应服务。 这些研究的结果将为公用事业运营商提供更有效的基础设施升级的见解，从而降低投资成本，并允许规划和分配现有资源，以及扩展和增强BlueWave-ai为公用事业提供的产品和服务。 该研究将通过计算模拟和BluWave-ai现有的分布式优化平台进行;开发的模型将使用Hydro渥太华的历史和网络数据进行验证。

The adoption of Electric Vehicles (EVs) and Photovoltaic (PV) generation at households is rapidly and massively increasing in the short term. Therefore, utilities are facing the challenge of efficiently manage these resources, to help mitigate the undesirable side effects of their operation when connected to power grids. Existing approaches to solve these issues rely on accurate predictions of the behavior of EVs and PVs (as part of Distributed Energy Resources or DERs), detailed knowledge of customers, and grid infrastructure, all of which complicate their effective deployment. Thus, the main objective of this project is to develop data-driven Artificial Intelligence (AI) based tools that remove the need of detailed modeling of certain parts of the distribution networks, and also address the issue of accurate forecasts. These tools, which will be developed to control DERs in distribution grids, will allow studying the impact of DERs and coordinating their operation to provide demand response services on utilities' networks. The results of these studies will provide insights for utility operators on more efficient infrastructure upgrades, thus reducing investment costs, and allow planning and allocation of existing resources, as well as expand and enhance the products and services that BlueWave-ai provides to utilities. The research will be carried out through computational simulations and BluWave-ai's existing distributed optimization platform; the developed models will be validated using historical and network data from Hydro Ottawa.