AI-Powered Energy Management for Fast Charging Hubs

We shape connector and site power in real time so your peak demand is lower while you still serve the same or more sessions. Our AI-DLM engine dynamically caps or shifts power using advanced machine learning algorithms to achieve the desired performance at all times. In practice you get fewer peak spikes, smoother load profiles, increased charging speed, happier drivers and lower monthly demand charges without touching your infrastructure.

Yes. There is an intrinsic inefficiency in power-limited fast-charging sites. By modeling the site, the drivers and the vehicles we are able to allocate power in a more optimal way, thus utilizing the grid power better and to the full extent.
This means that parallel sessions get higher average power and therefore dwell times are shorter and more sessions can be completed in an hour. The result is more kWh sold and more cars served per kW of grid capacity and much better driver experience.

Very simple. Before any commitment we run a digital twin simulation on your historical CDR data and site configuration. This 98% accurate simulation lets you see the impact on peak demand, demand charges, energy costs, throughput, charging speed, and more driver KPIs. You get a detailed before-versus-after view with clear estimates of savings and revenue uplift, and this allows you to calculate RoI clearly and easily.

The first thing that we do before activating our optimization engine is to accurately measure the existing driver experience (DX) metrics like dwell times, charging speeds, and a handful of additional DX-related proprietary metrics. Our driver-aware DLM is designed to improve these metrics over time and hence typical outcomes are higher charging speeds, fewer very slow sessions, fewer starved chargers etc.

Not at all. evPower.ai is vendor agnostic and works alongside your existing CMS/CPMS and charging hardware seamlessly. We use standard protocols and data points so you can keep your current chargers and backend 'as is'. code.

Tariffs, limits and constraints are configurations in our platform, not hard-coded logic. If your tariff structure or local rules change, we update the parameters and policies without touching your hardware. The optimization engine then adapts to the new economics so your sites keep operating at the best available profit and revenue profile.

No. We achieve very significant improvements without on-site storage. However, when an energy storage solution (BESS) is available, evPower.ai provides an additional layer of functionality that controls it (charge, discharge) to optimize its economic value.

No. This is a cloud-based software solution that integrates with your existing charging infrastructure and CMS/CPMS using standard protocols. There is no need for any hardware installation or changes to your current setup.
We do offer an industrial-grade local controller, that our customers use for on-site BTM (BESS, GenSet) control, or for local chargers control (over Modbus TCP, OCPP, or hybrid of both) in no-internet or high-availability scenarios.

No. evPower.ai is an energy management solution that works seamlessly alongside your existing CMS/CPMS and charging hardware.
For sites where no CPMS exists (e.g. private sites,fleets, etc..) we can operate in a standalone mode where we directly integrate with the chargers over OCPP or Modbus TCP and provide basic features, such as transaction analytics, charger fault notifications, and scheduling without the need for a full CMS/CPMS. In this case evPower.ai would be the primary control software for the site, but it is not a CMS/CPMS in the traditional sense as it does not provide features like customer management or billing that are typically associated with such platforms.

Yes. Our local controller can be easily integrated using common industrial protocols (Modbus RTC/TCP, DNP3, SNMPv3, MQTT) to control BESS charge and discharge, in addition to. Our techno-economic model for BESS combines both peak shaving and arbitrage methods in order to maximize RoI. This further reduces demand charges and energy costs by charging at optimal times and discharging during peaks.