CCLS wins GE Ecomagination Innovation Prize

The CCLS submission “Creation of a Columbia Machine Learning System to Optimize the Recharging of Electric Delivery Vehicles in Large Urban Cities” has been chosen as the “Winning University Program” in General Electric’s Ecomagination Innovation Challenge of 2010.GE’s CEO Jeff Immelt presented the award funding the 3 year, $1.1 million project, at a live event in New York City on November 16. Dr. Roger Anderson, PI of the CCLS proposal,  and representatives from collaborators Con Edison and Fedex Express were there to participate in one of the panel discussions.



The goal of this project is to build a Columbia Machine Learning System for intelligent Load and Source Management of EDV recharging facilities.  We will build a local intelligence system at the EDV depot that is also linked to the utility electric distribution control center and the onsite energy management system.

The world is struggling to find methods to reduce green house gas emissions, decrease dependence on hydrocarbon energy sources, enhance aging grid infrastructures, and improve living conditions. Electrifying the transportation sector has been heralded as a significant part of the overall solution. In dense urban areas such as New York City, delivery vehicles will hopefully constitute a significant portion of the electric transportation future. However, the switch to electrified charging depots for these fleet vehicles will cause localized stress points on the already heavily loaded distribution grid. Therefore, the use of electric delivery vehicles will drive the need for Smart Grid technologies to control this new and substantial load. Columbia University's Center for Computational Learning Systems will implement a Machine Learning System that will provide optimal recharging policies and actions that account for weather and other daily uncertainties connected to the distribution to empower urban Electric Delivery Vehicle (EDV) transformation. The MLS will recommend recharging strategies to a large number of advanced charge stations. These stations will have secure communication capabilities that will allow them to receive commands such as start and stop charging, while recording and then transmitting energy usage. More importantly, they will also be able to respond to load reduction directives to decrease or increase the current draw from the on-board vehicle inverter. The MLS will be connected not only to the recharge stations, but also to an advanced supervisory control and data acquisition (SCADA) system with integrated data historian. This system will serve as the charging depot’s central command center for vehicle charging. The MLS will use sophisticated algorithms to set the tuning parameters contained in SCADA’s control algorithms. The MLS system will also be connected to Con Edison’s distribution management system (DMS) responsible for overall electric grid distribution system operations. The utility can send signals to the MLS to alter the current or future load profile to ensure proper grid operations while also successfully meeting the vehicles’ energy demand constraints. Also, UPS battery power will be managed to assure reliability. This demonstration will quantify the realized, substantial benefits of a true smart transformation from oil-derived fuels to electricity for fleet delivery vehicles, resulting in tons of GHG per year being removed from the skies over NYC. In addition, EDVs will offer measurable noise reduction and improve public health by removing tuberculosis causing particulate emissions from the air.