This research presents a two-phase framework that predicts electric vehicle charging demand using location data and optimizes charger placement across the power grid to expand charging infrastructure without overloading it.
🚗💨 Electric vehicles (EVs) are on the rise, but one big question remains: where should we put all the chargers? Too few, and drivers face “range anxiety.” Too many in the wrong places, and the power grid might struggle. A new study from researchers at the University of Texas at Dallas takes on this challenge with a clever two-step strategy that blends data science, urban planning, and electrical engineering.
In this article, we’ll break down how the researchers combined Google Maps data, machine learning, and power grid simulations to design smarter charging infrastructure—tested in the Dallas–Fort Worth (DFW) area. ⚡🏙️
Building an electric vehicle charger network isn’t as simple as sprinkling chargers everywhere like Wi-Fi hotspots. Planners face two big challenges:
Most past studies looked at only one side of the problem: either where drivers need chargers OR what the grid can handle. This research blends both into a single framework.
The researchers developed a two-phase method:
👉 This phase gave a map of where chargers are needed most.
Once demand was mapped, the next step was figuring out where to actually install chargers without overloading the grid.
The researchers framed it as a “maximum coverage problem”:
They used a mathematical optimization model with factors like:
💲 Installation costs ($50k for new stations, $3k per charging port).
🚗 Travel distances for EV drivers.
⚡ Voltage and current stability in the 8500-node power grid.
The result: an optimized map showing where to add chargers, where to expand capacity, and how to keep the grid stable.
The team tested their framework in the DFW metroplex, one of the largest and fastest-growing areas in the U.S.
👉 In short: the framework worked in practice, not just theory.
For everyday EV owners, this research could mean:
🚫 Less “charging anxiety”—chargers are where you need them most.
⏳ Shorter wait times, since demand is matched with capacity.
⚡ Reliable charging, without surprise outages caused by grid stress.
Cities could roll this out to make EV adoption smoother, especially in fast-growing regions.
The study opens the door for more smart, data-driven EV infrastructure planning:
Scaling to Other Cities 🌎 This framework isn’t limited to Dallas. Any city with mobility data and grid maps could apply it.
Adapting for Fast Chargers 🚀 The study assumed Level 2 chargers (6 kW), common for homes and public lots. Future work could adapt the model for fast DC chargers, which demand much higher grid capacity.
Integration with Renewable Energy ☀️💨 Imagine pairing chargers with solar panels, batteries, or wind farms. Planning tools like this could ensure smooth integration without stressing the grid.
Dynamic, Real-Time Planning ⏱️ With connected cars and smart grids, cities might one day update charger deployment in real time based on traffic, demand, and power availability.
This research shows that building EV charging infrastructure doesn’t have to be a tug-of-war between driver convenience and grid stability. By combining machine learning, urban data, and electrical engineering, cities can future-proof their charging networks. 🚙⚡
As EV adoption surges, smarter planning like this will make the difference between a patchy charger map and a seamless, reliable charging experience.
🔌 Electric Vehicle (EV) - A car that runs on electricity stored in batteries instead of gasoline. Think of it as your phone on wheels—plug it in, charge it, and go! - More about this concept in the article "Charging Up the Future ⚡️ Predicting EV Fast-Charger Demand on Motorways with Smart Simulations 🚗🔋".
⚡ EV Charger - The device that “refuels” an EV by supplying electricity to its battery. Chargers range from Level 1 (slow, like overnight charging at home) to DC Fast Chargers (super quick, highway pit stops).
🏙️ Charging Infrastructure - All the physical and digital systems that make EV charging possible—stations, chargers, software, and the power connections behind them. Basically, the EV world’s version of gas stations.
📍 Points of Interest (POIs) - Places people frequently visit, like malls, schools, grocery stores, and cafes. These spots are great candidates for charging stations because drivers already stop there.
🧮 Machine Learning (ML) - A branch of artificial intelligence where computers learn patterns from data and make predictions—like guessing how many chargers a grocery store parking lot might need based on visitor numbers. - More about this concept in the article "Smarter, Stable Smart Grids ⚡ Hybrid AI".
🌐 Power Grid - The huge network of wires, transformers, and substations that delivers electricity from power plants to your home, office, or an EV charger. It’s like the internet, but for electricity.
🚌 Distribution Grid - The local “neighborhood level” of the power grid that actually connects to homes, businesses, and EV chargers. If the power grid is the highway system, the distribution grid is your local streets.
🧩 Maximum Coverage Problem - A mathematical puzzle that finds the best way to cover the most demand with the least resources. In this case: where to place chargers so the most drivers are served with the fewest stations.
📊 XGBoost (Extreme Gradient Boosting) - A popular machine learning algorithm that’s great at making predictions. Here, it helps forecast EV charger demand based on city data. - More about this concept in the article "Smart Trains, Greener Cities 🚆 How AI-Optimized Scheduling Slashes Carbon Emissions in Hangzhou 🌍".
Source: Harshal D. Kaushik, Jingbo Wang, Roshni Anna Jacob, Jie Zhang. Electric Vehicle Charger Infrastructure Planning: Demand Estimation, Coverage Optimization Over an Integrated Power Grid. https://doi.org/10.48550/arXiv.2509.23699