For this Data Science Insight I read an article written by Junjie Hu, Hugo Morais, Morten Lind, Henrik W. Bindner, published in an Electric Power Systems Research. Titled: ‘Multi-agent based modeling for electric vehicle integration in adistribution network operation’

The purpose of this article is to outline a method for managing excessive energy consumption. Many other articles published before and after this one (2016) describe how to acheive efficient management of energy, but this article chooses to specify the issue to meet the increase in the amount of electric vehicles.

The “paper aims to develop an agent-based control sys-tem for simulating and operating future distribution networks characterized by high penetration of demand side resources”. Hu et al. identify thermal capabilities of cables and power transformer limits as some critical challenges for distribution networks with congestion problems. Below are diagrams pulled from the article that outline the agent-based model and the information flow from one node to the next.

There are four agents used in this model. The DSO Tech agent (Distribution sytem operator technical agent) manages energy demands and reduces the negative effects of high congestion. The DSO Market agent ascertains the level of congestion in relation to the capacity of the transformer and sends data to the EV VPP agents, which “manages conventional loads and the electric vehicle charge process”. This method will adjust the electric vehicle’s charging schedule to meet current energy consumption needs identified in the model-user’s data. One caveat is that each owner of an electric vehicle must agree to send data concerning trip requirements and battery status to the VPP agents for this method to work at scale. Considering the ease-of-access companies generally have when acquiring rights to user data, this should not be a problem for major energy companies who already have access to ungodly amounts of resources and legal representatives. The software used in the brokerage between each of these four types of agents is simple, and is expressed in the following diagram pulled from the article:

This topic is interesting as the concept is ahead of its time. Soon enough, the transition towards sustainable forms of energy use will become more of a necessity than an option. This year the Governor of California, Governor Newsom, pledged to phase out gasoline powered cars. Additionally, there are multiple similar mandates in the state to update sustainable infrastructure and promote increased use of fully-electric vehicles that have already been added to the legislature. As the increasing threat of climate change is ignored by our leaders, the jump in system congestion and the need for re-shaping of energy architecture will only be more dramatic once we decide it is time to face this threat. Thus, I believe there is a legitimate need for this type of application of Agent-Based modeling in the energy sector.