The electricity market is writing a new script. And any operator of a utility-scale solar farm who still assumes that generation volume alone carries the business case is in for a rude awakening. At an engaging seminar organised by pv magazine, expert Filip Markert of e2m spoke about price developments in the electricity market.
The numbers speak for themselves. The monthly market value for solar power currently stands at just 1.3 cents per kilowatt hour, as calculated by the direct marketing experts at e2m in Leipzig. The annual market value, meaning the weighted average price a PV plant achieves over a full year, has also come under massive pressure: while solar generation still fetched around 76 per cent of the average exchange price in 2023, it is now down to roughly 50 per cent. A drop that shows up directly in operating results.
The cause lies in the physics of the electricity market. There are more and more days on which photovoltaics alone accounts for Germany's power supply. The consequence: massive oversupply at the midday and afternoon peak. Precisely when the sun shines hardest and the most solar power flows into the grid, the day-ahead market price collapses. On particularly sunny days it even turns negative.
Utility-scale battery storage can absorb electricity in the megawatt range and release it again later. That increases the achievable revenue on the power exchanges.
This is the central problem. A utility-scale PV plant produces a lot exactly when the electricity market is oversupplied, and it is penalised for that. Conversely, prices are high in the evening, when the sun sets and demand rises. That is where the economic potential lies which has so far gone unused.
Flexibility and battery storage as the new secret weapons
New utility-scale solar farms are therefore no longer planned as pure generation assets. The maxim “no new utility-scale PV without storage” has become the industry standard, and for good reason. The same logic should be applied to existing plants.
Large-scale battery storage (BESS) turns the problem into an opportunity: at midday, the solar plant does not feed directly into the grid but charges a battery. That battery can release the electricity later, at better prices, typically in the evening, when prices are usually three to four times higher than during the day.
Technically this is feasible. Intelligent control systems, so-called co-location setups, use real-time price signals from EPEX. The battery is charged when the spot market price is low or turns negative. It is discharged when prices pick up. The state of charge is continuously reconciled with intraday prices and the weather forecast.
The result is impressive. According to practical examples from recent market analyses, co-location systems can increase revenue potential by up to 250 per cent, not through more production but through better marketing. The key is not the kilowatt hour itself, but the right moment to sell it.
From power plant to trading instrument
A profound paradigm shift becomes visible here. The business model is no longer a “PV plant case” as classic developers understand it. It is a “trading storage case”. Photovoltaics becomes the raw material supplier, the battery becomes the active market participant.
This also means that allocation decisions between production and storage turn into complex planning horizons. Charge or not charge? Release now or wait? These decisions depend on dozens of variables: state of charge, weather forecast, intraday price curves, balancing energy prices, and often even the Federal Network Agency's forecast for the next day.
The alternative to this active control is so-called spot market participation with curtailment. The operator gives up the market premium and receives the spot market price instead, but is automatically switched off during price extremes. That reduces overproduction in price troughs and leads to better average values. Admittedly less spectacular than a 250 per cent revenue increase through co-location, but considerably easier to manage.
The decisive point is that this shift is not optional. Anyone operating utility-scale PV purely as a conventional generation asset will increasingly be pushed to the edge of economic viability. The spreads between midday and evening prices are widening, overcapacity is growing, and monthly market values keep falling.
The technical foundation: vertical data integration
For this model to work in practice, however, more than a smart battery is needed. The complexity of decisions such as “charge or not charge”, “discharge at the midday peak or in the evening” and “forecasting production peaks” requires a data platform that brings all sources together in real time: exchange prices from EPEX, weather forecasts, battery telemetry, grid frequencies.
Only with vertical data integration do these raw signals become a basis for decisions on which automated trading logic can rest. At the same time, predictive maintenance on the battery enables proactive optimisation instead of reactive failure management. That is the difference between intelligent co-location and wishful thinking.
For a long time, battery storage was the answer to grid stability and security of supply. It is now becoming the answer to flexibility as a pure-play business model. And for existing plants it is a second lease of life, provided operators and developers understand and accept the new rules of the game.
The key terms at a glance
BESS (battery energy storage system)
Utility-scale battery storage that can absorb electricity in the megawatt range and release it again later. In a PV context, it is the central element for decoupling generation from sale in time.
Co-location
The combination of a generation asset and battery storage at one site, usually behind a shared grid connection point. The advantages are shared infrastructure, grid connection and control. Only joint operational management turns two assets into one marketable system.
Direct marketing
Selling the generated electricity on the exchange through a direct marketer rather than under a fixed feed-in tariff. The operator receives the exchange revenue and, where support applies, the market premium as the difference to the applicable value.
Market value (monthly and annual)
The volume-weighted average exchange price a technology actually achieves in a month or a year. Because solar plants all feed in at the same time, the solar market value is systematically below the general average price. It is also the reference figure used to calculate the market premium.
Day-ahead market
Trading for the following day, settled in a daily auction with hourly and quarter-hourly products. The price formed here is the central reference for revenues, market values and negative price hours.
Spot market and intraday trading
The spot market covers short-term electricity trading, that is, day-ahead and intraday. In intraday trading, continuous trading runs until a few minutes before delivery. It is precisely this window that storage uses to react to forecast deviations and price jumps.
EPEX SPOT
The European power exchange for short-term trading, which includes the German market area. Its price signals are the input for every automated charging and discharging decision.
Data integration
Bringing together data from different sources into one consistent, time-synchronised model. “Vertical” means from the plant's sensors through to the market signal. Without this layer, price, weather and asset data remain separate silos and cannot be traded on.
Predictive maintenance
Maintenance based on condition and telemetry data, and on models trained on it, that predict a failure before it occurs. For battery storage this concerns degradation, thermal behaviour and cell condition in particular, exactly the variables that determine availability for trading.
About control-f. control-f GmbH is a values-driven AI company based in Konstanz. Since 2022, the data boutique has been building big data platforms for industrial telemetry data and helping companies across the German-speaking region make complex data landscapes usable. Its clients include large corporations and mid-sized companies in plant engineering, automotive and the energy industry. Managing directors Simon Deussen (machine learning engineer and founder) and Daniel Tremer (formerly Specialist Data Science & AI Projects at Porsche AG) focus on building stable data architectures as the foundation for analytics, software solutions and AI applications such as predictive maintenance.
Kategorie / Category: Blogposts