Microgrids

TU/e
The campus
self-optimizing

Battery
Max charging capacity
1500 kW
Max capacity
3500 kWh
Chargers
Max charging capacity
20 x 22 kW AC
 
528 kW
Connection
Physical
10000 kW
Contracted supply
- kW
Contracted feed-in
- kW
Operator
Operator
Zympler
Hardware
Delta
Energy supplier
Powerhive
Gateway
Zympler

The situation

A complex campus topology with fragmented charging stations, solar panels, and heat pumps spread across multiple subnets and transformers falls under a single Administratively Consolidated Connection (ASA), which is consistently exceeded on cold winter days.

A battery used solely for peak shaving therefore leaves significant value untapped throughout the year.

Metering points
68
Additional connection capacity
1000 kW
Battery payback period
7 years

Key challenges

Fragmented infrastructure, one connection limit

Batteries, charging stations, and solar panels spread across the campus, all under one shared ASA connection that is exceeded on cold winter days when the heat pumps kick in.

Peak shaving without idling the battery

Most systems keep the battery at 100 percent State of Charge all year round. That's safe but a waste. The challenge was to look ahead, charge smartly, and simultaneously profit from the energy market.

Operationally reliable and open to research

The battery had to supply the campus with energy AND be available as a test platform for TU Eindhoven's Deep Reinforcement Learning research, without these two goals interfering with each other.

Thijs Meulen
Advisor, Energy Management and Building Automation, TU Eindhoven

"While other EMSs let the battery sit idle all year, Zympler keeps it working all year."

TU Eindhoven and Zympler

Predictive peak shaving

Zympler predicts when heat pumps will generate additional demand and ensures the battery is fully charged at the best prices during those times, without unnecessarily keeping it at 100% charge throughout the year.

Trading outside peak windows

Outside of predicted peak periods, Zympler actively traded the battery on the electricity markets. This makes the business case realistic, with a payback period of 7 years.

Open platform for AI research

In collaboration with TU Eindhoven, during periods without peak shaving demand, the battery is made available for students and researchers to test Deep Reinforcement Learning algorithms in a real-world environment.

EIRES & Research Base

The Eindhoven Institute for Renewable Energy Systems (EIRES) supported the project with scientific insights from energy research, connecting system innovation, scientific research, and societal impact.

Why are Zympler and TU Eindhoven a good match?

The TU Eindhoven campus combines technical complexity with research ambitions, and Zympler delivers exactly that: an EMS that not only manages the complex topology but also provides the space to use the battery as a living laboratory. Both parties learn from each other, the campus recoups the battery's cost, and science benefits from a real-world testing environment.

  With Zympler, TU Eindhoven manages a hyper-smart energy grid.

Relevant cases

Boxtom

Public charging with smart battery control

TU/e campus

The self-optimizing campus

50five

Silent batteries at work

R. Nagel Transport

Smart charging at the grid edge

Want to know more about how Zympler and TUe innovate?