What is the value of a battery storage system in electricity trading? Start-up company Re-Twin Energy developed an AI-powered software that provides a transparent, consistent and project-specific answer to this question.
Digital twins are used to create realistic models of battery storage systems and simulate their performance in the electricity market, enabling the proper assessment of profitability, risks and operating strategies throughout their entire lifecycle. Learn more about Re-Twin Energy’s platform in this interview with Florian Heise, the start-up’s co-founder.
Mayur Andulkar, CEO of Re-Twin Energy, and I founded the company about a year ago. Prior to that, we worked for a green hydrogen developer for about two and a half years.
Our job at our former company was to develop economically viable hydrogen projects. This was no easy task. Given the currently high costs of green hydrogen, we explored additional revenue streams. When examining a technical setup with battery storage systems, we came up with the idea of using them in electricity trading to generate maximum profit. To do this, we also had to determine the financial value of battery storage systems in electricity trading.
This analysis was complex and time-consuming. The answers we got from traders offered little transparency, and expensive consultancies only supplied static data. However, the results showed that battery storage systems could be marketed much faster and more reliably than hydrogen projects.
Armed with this knowledge, we developed Re-Twin Energy, a software solution that enables users to create digital twins of their battery storage systems and calculate their value in electricity trading.
The trader-grade AI model is the primary AI analysis model used in the Re-Twin Energy platform. It links the digital twin of a battery storage system with market data and optimizes its financial strategy accordingly. The digital twin models the plant – including all its technical and operational properties – while the AI model interprets market mechanics and derives the ideal trading strategy for the analysis.
Virtual trading allows clients to simulate trading electricity from their storage systems without participating in the actual market. The goal is to reveal the potential revenue that could be generated by a given plant and strategy. All assumptions, steps and results are disclosed.
The virtual trading provides a daily benchmark data point. Clients receive a simulated trading plan for the following day, along with information on how much revenue the plan would have generated. This value can be directly compared with the actual performance of a commissioned trader. The benchmark defines a minimum performance level that traders should achieve, establishing an objective basis for performance assessment.
Our platform primarily answers the question of how much profit a battery storage system can generate in electricity trading. The answer depends on many parameters, including technical aspects such as configuration and technology, site-specific factors such as weather conditions, grid connection and grid availability, as well as market-based decisions on trading strategies and products. Add to that the operator’s individual risk appetite, which has a major impact on the trading strategy.
It is hard to keep track of and manage this combination of different parameters. That is why our AI-based model integrates all relevant data to provide project-specific, reliable statements instead of general estimates.
Another advantage of our software is that it takes project dynamics into account. Assumptions about a project change throughout the planning, investment decision and operational phases. Our software can quickly integrate these changes and recalculate the financial value of the storage system without requiring a manual update of the business case.
The primary goal of developers, operators and investors is to determine which storage projects are worth investing in and whether the proposed design is economically viable. It is true that the available assessments are often based on fragmented analyses. This is partially due to long-term revenue forecasts from consultancies that are based on current assumptions and provide a snapshot of the situation today and in the distant future. It is also due to traders providing historical revenue data from similar projects based on their own opaque methods. Banks then have to combine these different approaches to create high, base and low case scenarios.
Our platform replaces this fragmentation with a unified methodology and consistent calculation logic across all time horizons. It links a long-term business case, historical backtests based on real market data, and continuous market simulation via digital twins. This makes it transparent what revenues a storage asset would have generated in the past, what revenues are realistically achievable in the future, and what revenues are currently being generated – regardless of the asset’s stage of development. As a result, customers receive a holistic, practical, and comparable assessment of economic performance without having to reconcile conflicting assumptions from different sources.
For instance, Re-Twin Energy’s AI services help utilities assess and prioritize their project pipeline. A typical application scenario is determining which existing PV systems would benefit financially from the addition of battery storage systems.
For a client project we received last-minute information that the grid connection is asymmetrically limited, with different capacities for feeding into and drawing from the grid. Our tool was able to incorporate this change immediately to create a new scenario. As a result, the business case worsened significantly, causing the project to fall below the internal target revenue.
Our tool allows project teams to swiftly assess such changes and use their findings to make well-informed decisions, ensuring more robust investment decisions and avoiding misjudgments early on.
The main difference lies in the seamless use of a digital twin throughout the battery storage system’s entire lifecycle. Today, many storage projects are still planned using only conventional analyses. After several years of operation, what is often missing is a solid, data-driven foundation for systematically assessing changing market conditions, ageing effects and operational decisions.
With Re-Twin Energy, the digital twin is first used in the planning phase for scenario analyses, and later runs alongside the battery’s actual operation as a live simulation. In the long term, the digital twin allows for risk-free experimentation, such as evaluating alternative operating strategies or simulating extreme weather conditions, without interfering with the actual plant. After several years of operation, this ensures a much higher transparency, risk minimization and decision quality than can be achieved with purely conventional approaches.