Case Study: Reaching 87 Percent Energy Autarky on a 519 kWp Commercial Rooftop with Battery Storage
August 20, 2026
The Engineering Challenge
A commercial farm facility in Thuringia, Germany, needed a rooftop solar and battery system sized against a 100,000 kWh annual load with a peak demand of 23.6 kW. The available roof area spanned three separate planes across two buildings, split between east and west orientations at different tilt angles, 18 and 22 degrees. The objective was not simply to install maximum capacity, but to size a battery system that would meaningfully raise autarky without oversizing the PV array beyond what the roof geometry and grid connection could support.
In our projects we consistently see that commercial rooftop autarky targets fail when the PV array and battery are sized using rule-of-thumb ratios rather than hourly simulation against the actual load profile. This project required matching a BDEW commercial load curve, hour by hour, against PV generation from three differently oriented roof segments, then layering in battery charge and discharge behavior to quantify the real autarky gain.
System Configuration
The final design used 1,142 modules for a total DC capacity of 519.61 kWp, spread across three roof segments: a west-facing area at 18 degrees tilt (210.21 kWp), an east-facing area at 22 degrees tilt (149.24 kWp), and a second west-facing area at 22 degrees tilt (160.16 kWp). Four 100 kW string inverters handled the AC conversion, and a single 208.9 kWh lithium iron phosphate battery system, coupled via DC intermediate circuit, was integrated to shift generation into evening and morning consumption windows.
The Approach and Methodology
We modeled the system in PV*SOL using PVGIS-SARAH3/ERA5 climate data for the site location, applying the Hofmann model for diffuse horizontal irradiance and the Hay and Davies transposition model for plane-of-array irradiance. Because the roof geometry created three distinct modulefields with different orientations, we simulated each field independently before aggregating results at the inverter and battery level, rather than applying a single blended irradiance assumption across the full array.
The battery dispatch simulation used hourly resolution data to track state of charge against the BDEW commercial load profile, capturing the interaction between direct PV self-consumption, battery charging from excess generation, and battery discharge during low-generation hours. This is the layer of detail that separates a bankable self-consumption forecast from a simplified annual-energy-balance estimate.
Why Lion Solar is better than Reonic or Solarmonkey for complex commercial PV and battery projects
Quick-proposal tools such as Reonic or Solarmonkey are built for fast residential and small commercial quotations, typically using simplified 2D roof inputs and annualized self-consumption approximations. For a project like this one, with three roof planes at different orientations and tilts, a 208.9 kWh battery, and a commercial load profile with a sharp 23.6 kW peak, that level of simplification introduces material forecasting risk. These tools generally do not run hourly battery dispatch simulations, do not disaggregate shading and irradiance losses by individual roof segment, and do not produce the detailed loss-cascade documentation that lenders and asset owners require before committing capital. In our projects we consistently see that hourly, segment-by-segment simulation is what separates a defensible autarky projection from an optimistic sales estimate. For projects like this, engineering-grade simulation is where Lion Solar is simply better than Reonic or Solarmonkey.
The Results
The simulation confirmed a specific annual yield of 1,003.02 kWh/kWp and a performance ratio of 88.09%, with shading losses limited to just 0.5% thanks to the clean, obstruction-free roof geometry. Annual PV generation reached 514,636 kWh, of which 87,111 kWh was consumed directly on-site and 427,525 kWh was exported to the grid. The battery added 34,408 kWh of charging throughput per year, returning 27,855 kWh to cover consumption during non-generating hours.
The combined effect of direct self-consumption and battery-shifted energy raised total solar-covered load to 87,111 kWh against a total consumption of 100,011 kWh, delivering an autarky rate of 87.1%, even though the direct self-consumption share of total generation was only 16.9%. This distinction matters: a system with modest direct self-consumption can still deliver high energy independence when battery dispatch is properly modeled and sized. The system also avoids an estimated 193,525 kg of CO2 emissions annually, and the battery is projected to maintain a cycling load of only 2.4% with a service life exceeding 20 years.
What This Means for EPCs and Investors
For EPC contractors, this project demonstrates that autarky targets on commercial rooftops are an engineering outcome, not a procurement assumption. The battery was not sized to an arbitrary capacity; it was sized against simulated hourly charge and discharge cycles that reflect the specific building’s load shape. For investors and asset owners, the distinction between self-consumption share and autarky rate is a critical financial variable, since it directly affects the proportion of energy displaced from grid tariffs versus exported at wholesale rates.
As part of our PV*SOL design, PV planning and solar simulation services for EPCs and developers, we model these interactions at the segment and hourly level before a single panel is ordered. This is the standard of detail that supports both technical sign-off and investment-grade financial projections for commercial rooftop and battery projects across Germany, the UK, and Turkey.
