Glossary

What changes for peak shaving: today a single unfavorable 15-minute value in the year determines the demand charge. Under AgNes three components take its place — a capacity price for capacity booked in advance, working price 1 on every kilowatt-hour within that booking and working price 2 as a penalty price on every kilowatt-hour above it. Peak shaving then no longer means shaving one annual peak, but staying below the booked capacity permanently. Also planned are location-dependent dynamic grid fees and, for the first time, fees on feed-in.

In practice that means: instead of accepting the worst-case blanket assumption, a metering concept proves how much grid power actually flowed into the storage. It costs metering effort, but it pays off when the grid power share is small.

The logic behind it: the grid is built out for the moment of highest total load. Whoever draws less at exactly that moment relieves the grid and pays less. That is why only the reduction inside the high-load windows counts, not the annual peak load. Your distribution network operator defines the windows and publishes them.

Rule of thumb: BTM earns on costs that never arise, FTM on revenue that does. A BTM storage competes with an electricity price of maybe 25 ct/kWh, an FTM storage with a spot price of maybe 8 ct/kWh. That is why storage behind the meter often pays off at much smaller sizes.

What changes in the result: by default Furo compares a load profile with PV but without a battery against one with PV and a battery, so the PV system itself doesn't show up in the result. In the combined analysis the PV investment cost and the PV savings are added, plus three extra tables: PV electricity savings, PV payback and PV feed-in revenue. Requirements: at least one PV system of plant type “Planned” and PV investment cost greater than zero. There is no separate payback for the storage alone in the combined analysis.

FTM projects are larger, regulatorily more demanding and depend entirely on market prices, because there is no consumer behind them whose bill they could lower.

The difference is not physical but an accounting one. Power in the grid can't be separated by origin; what counts is what is metered and allocated. If the storage charges PV power only, the subsidy is preserved. As soon as grid power flows in as well, it counts as mixed mode, and it has to be proven which share came from where. That is exactly what the allocation option (Abgrenzungsoption) is for.

Grid power is not inherently bad for the economics. Charging cheaply at night and shaving an expensive peak load during the day can pay off even with levies on top. The catch is the subsidy, not the electricity price. That is why Furo asks whether the storage may charge grid power, and factors in the consequences.

The sales argument behind it: a battery that only optimizes self-consumption earns money only when the sun happens to be shining or a peak load is coming. The rest of the time it sits idle. Marketing means turning that idle time into money. Two forms: direct marketing sells generated PV power, flexibility marketing sells the battery's ability to charge or discharge on signal.

Background: storage that charges PV power only is regulatorily simple. As soon as it also charges grid power, subsidized and unsubsidized electricity mix, and it has to be settled which share carries which levies. MiSpeL governs exactly this mixed mode.

In one sentence: what are all future savings worth once you convert them to today, minus what you invest today? If the value is positive, the project earns more than putting the money elsewhere. It is discounted because €1,000 in ten years is worth less than €1,000 today.

Careful in sales: the highest net present value is not the same as the best offer. Larger batteries often have the higher absolute net present value, but they tie up more equity and pay back more slowly. One or two sizes smaller is frequently easier to sell. So show several sizes side by side instead of just the recommendation.

Important for the customer conversation: the streams don't simply add up. The battery can only use a kilowatt-hour once. If it charges cheaply for the spot market, the capacity may be missing for the afternoon peak load. That is why the split shifts as soon as settings change, and why the total comes out lower than the sum of the use cases calculated individually.

What matters for marketing is not the nominal capacity but the free headroom. A 1,000 kWh battery operating between 40 and 60% has 200 kWh free. The same battery between 20 and 80% has 600 kWh free. That is why the arbitrage revenue varies from day to day.

The idea is simple, the execution is not: the price difference between the cheapest and the most expensive hour has to be larger than the efficiency losses plus the levies on the charged power. Of a given price spread, often less than half is left as revenue.

The picture behind it: a fixed band of power is bought in advance, say 75% of annual consumption at a price fixed today. Anything above it is bought on the spot market, anything below it is credited. For the battery that means only the variable share can be optimized; the fixed share is paid for regardless of when you consume it.

Example: 1,000,000 kWh of annual consumption at a 400 kW peak gives 2,500 usage hours. Below the threshold you get a lower demand charge (Leistungspreis) and a higher working price, above it the other way round. A battery lowers the peak and therefore shifts operation upwards, which can be cheaper or more expensive depending on the consumption profile. This is exactly where the seemingly contradictory results come from.