Open, data-driven tools for smart electronic vehicle charging

The safe and affordable integration of millions of electric vehicles (EVs) into the grid will require advanced scheduling algorithms as well as techniques to optimize charging facilities. However, a lack of data, simulation environments and physical testbeds have hampered research in these areas.

To address this gap, we have developed a new collection of data and tools called the Adaptive Charging Network (ACN) Research Portal. This portal includes ACN-Data, a publicly accessible dataset of real charging sessions, ACN-Sim, an open-source data-driven simulation environment for evaluating EV charging algorithms and facility designs and ACN-Live, a testbed for field testing online algorithms by controlling real charging stations.

This talk will describe the ACN Research Portal and the research questions it enables. In particular, we focus on two applications relating to large-scale charging facilities such as those on universities, workplaces or apartment complexes. First, we consider the problem of scheduling EV charging to reduce operating costs, including time-of-use rates and demand charges, while respecting infrastructure constraints.

We propose a model predictive control framework and compare it with conventional deadline scheduling approaches using real workloads from ACN-Data and realistic models from ACN-Sim. Second, we consider the planning problem of optimally sizing on-site PV generation to minimize capital and operating costs of an EV charging facility. To do this, we propose a scenario-based optimization to minimize total expected cost based on statistical models learned from ACN-Data and evaluate our approach using ACN-Sim.


This is a virtual seminar, please join us via Zoom





Zachary Lee is a Resnick Sustainability Institute Fellow at the California Institute of Technology. He is a PhD student in electrical engineering advised by Professor Steven Low. He earned his BS Eng. from John Brown University in 2016, and his MS from Caltech in 2018. Zach's research is centred on smart electric vehicle charging, where he draws on broad interests at the intersection of data science, computer science and energy. 

Zach helped develop the software and algorithms that make up the Adaptive Charging Network (ACN), a framework for smart charging systems, which was developed at Caltech and has since been commercialized by PowerFlex, a Caltech startup. More recently, he has focused on distilling the data and insights captured from building actual charging systems into an open-source toolbox for EV charging research. As a lover of travel and of the outdoors, he hopes that affordable and ubiquitous charging infrastructure will speed the adoption of electric vehicles worldwide, allowing us to be good stewards of our environment without restricting mobility.

Date & time

11–11.50am 19 May 2020



Zachary Lee

Updated:  10 August 2021/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing