BEMS

BEMS stands for ‘Balcony Power Plant Energy Management System in Low-Data Environments’. The research project is developing smart energy management systems for small photovoltaic installations (balcony power stations) to optimise the self-consumption of solar power – either with a focus on cost-effectiveness or to maximise self-sufficiency. The focus is on the use of artificial intelligence (AI) on cost-effective embedded hardware that can make reliable decisions even without extensive historical data. The project is funded as part of the Central Innovation Programme for SMEs (ZIM) of the Federal Ministry for Economic Affairs and Energy (BMWE) and is being implemented in collaboration with the industry partner fothermo System AG.

Institute for Communication Technology (IKT) 

Why BEMS?

Balcony power plants enable private households to meet part of their electricity needs themselves. However, since electricity generation depends on the weather and the time of day, and household energy consumption fluctuates, energy surpluses often occur. This surplus is usually fed into the public power grid with little or no compensation, meaning the economic potential of these systems is often not fully realized.

BEMS aims to intelligently distribute this energy—and, whenever possible, use it within the household itself—by integrating electrical and thermal energy storage systems. This improves both the economic efficiency of balcony power plants and their contribution to the energy transition.

 

What makes BEMS special?

A key focus of the project is the development of AI-based algorithms for data-poor environments. In contrast to traditional energy management systems, there is often little or no historical consumption and generation data available at the time of commissioning. BEMS is therefore developing intelligent algorithms that make energy-optimized decisions even under these conditions and continuously adapt to users’ consumption patterns.

The AI algorithms developed are executed directly on cost-effective embedded hardware. This allows energy flows to be processed locally, quickly, and in a privacy-friendly manner without relying on a permanent cloud connection.

 

Project Focus
  • Development of an AI-based energy management system for balcony power plants
  • Optimization of self-consumption of photovoltaic energy
  • Intelligent control of electrical and thermal storage systems
  • Development of resource-efficient AI methods on cost-effective embedded hardware
  • Development of adaptive algorithms for low-data environments
  • Easy retrofitting of existing balcony power plants

 

Application Examples

 

  • Smart Storage Control

    The battery storage system is charged with excess solar power or during periods of favorable dynamic electricity rates, rather than feeding excess energy into the public grid at a low rate.

  • Optimized Energy Use

    The hot water temperature is regulated for optimal energy efficiency, taking into account user comfort, heat loss, and hygiene requirements such as limescale and Legionella growth.

  • Adaptive Energy Management

    The system learns users’ individual consumption patterns and preferences and continuously adjusts the distribution of energy flows accordingly.

 

Our Team

The BEMS project is being carried out by a team at the Institute of Information and Communication Technology (IKT) at Ulm University of Applied Sciences. The researchers are developing innovative methods in the fields of AI, embedded systems, and energy management to create smart, cost-effective, and easily retrofittable solutions for the use of balcony power plants in private households.

The project is being carried out in close collaboration with Fothermo System AG. The company is developing a plug-and-play heating element for hot water production that can be easily integrated into existing systems and specifically converts excess photovoltaic power into thermal energy. This technology serves as a practical platform for developing and testing the energy management methods developed in the project.

Professor
Faculty Electrical Engineering and Information Technology
Placeholder