Algorithms for the energy industry

Fraunhofer SCAI develops tailored algorithms for the energy industry to meet the sector’s evolving requirements. The work is grounded in advanced mathematical methods and modern software engineering. This enables customers and partners to address use cases ranging from efficient planning and optimization of energy networks to the sustainable valuation and risk management of physical assets.

Strengths

At Fraunhofer SCAI, scientific excellence meets industrial practice. Research results have been translated into operational solutions for many years. In collaborative industry projects, state-of-the-art research is combined with proven implementation expertise. Core capabilities include optimization, high-performance numerical solvers, simulation techniques, and artificial intelligence. Reliability and long-term partnership are key principles. Customer data remains strictly confidential and is handled with the utmost care.

A partner for intelligent energy solutions

SCAI's combination of interdisciplinary expertise and practical implementation delivers technically robust, economically viable solutions. Companies and public institutions benefit from clear, actionable answers to complex energy-system questions. SCAI distinguishes itself by delivering tailored algorithms as a trusted, long-term partner.

Software solutions and application areas

 

Energy network planning

  • MYNTS: software for planning network infrastructure for gas, electricity, hydrogen, and CO2
  • Simulation and planning of physical sector coupling
  • Assessment of network resilience and flexibility
  • Cost analyses
 

Renewable price simulation

Fraunhofer SCAI develops stochastic models to simulate prices influenced by volatile renewable generation. These models support dynamic hedging and valuation strategies. Established approaches, such as least-squares Monte Carlo, are used and complemented by modern machine-learning methods. The results help energy companies and investors make more robust trading and investment decisions under uncertainty.

 

Energy portfolio optimization

The software “EAO – Energy Asset Optimization” optimizes heterogeneous energy portfolios in a structured, value-driven way. It reduces risk and improves returns. EAO models wind and solar assets, battery storage, gas-fired power plants, industrial demand, and thermal storage within a single portfolio. Municipal utilities, energy suppliers, trading houses, and investors use EAO to value and manage their assets.

 

Trading strategies for Green PPAs

Green power purchase agreements (Green PPAs) offer new opportunities but also introduce material price and weather risks. Since weather data is not tradable, effective hedging and risk management are essential. Self-supervised learning and deep hedging are applied to develop optimized trading strategies for portfolio management. This enables more resilient PPA structuring and operation.

 

SCAI Battery Value Indicator

The SCAI Battery Value Indicator (SBVI) provides a transparent benchmark for assessing short-term revenue opportunities for battery storage assets. They are computed daily to estimate the value that could have been earned on the previous day, reflecting revenues across the day-ahead, intraday auction, and intraday markets while accounting for liquidity effects through bid–ask spreads in intraday trading.

 

New materials for energy storage

Renewables reach their full potential when paired with energy storage. Today’s storage technologies often rely on materials associated with critical raw materials. Innovative algorithms accelerate the identification of alternative materials.

 

Predictive modeling of battery aging

Machine learning can significantly accelerate battery aging simulations by guiding the solution of linear systems. This enables the use of highly efficient numerical solvers without compromising robustness. The approach reduces computational time and improves model accuracy. It also supports faster design iterations and more reliable lifetime predictions in research and industrial applications.

Semantic information management

Semantic data modeling links information from different sources in a consistent, unambiguous way, enabling automated processing. This increases efficiency in grid operations, market communication, and asset management. It lowers costs and accelerates decision-making. Examples include: GEAR-UP, BASE, SmartEM, RESTORE, ALABAMA, Pioneer, ITEA VMAP analytics

 

Fast oil and gas reservoir simulation

The SAMG reservoir interface accelerates subsurface simulations and improves numerical robustness for models with multiple phases and components, thermal effects, and coupled geomechanics. The same approach also applies to geothermal projects, gas storage, and CO2 sequestration (CCS). This supports larger, more detailed models and accelerates workflows and studies.

 

Rapid simulations with SAMG-Modflow

SAMG-Modflow accelerates groundwater and surface-water simulations using an efficient algebraic multigrid solver. Machine learning reduces parameter tuning, while multithreading leverages multiple CPU cores. This shortens runtimes for infrastructure, hydropower, and transport models. It also maintains numerical robustness for large-scale models.