Contents
- Distributed generation installations statistics: U.S. distributed solar system sizes
- Distributed generation installations statistics: technology in residential and non-residential systems
- Distributed generation installations statistics: storage pairing and battery sizes
- Distributed generation installations statistics: mounting and panel orientation
- Distributed generation installations statistics: reported regional installation totals
- Sources
Distributed generation installations statistics: U.S. distributed solar system sizes
For U.S. residential distributed photovoltaic (PV) systems installed in 2024, the median system size was 7.2 kW and the average was 9.3 kW. These figures describe two different points in the size distribution: the median marks the midpoint, while the average reflects the full set of recorded sizes. The difference between them indicates that the average was higher than the median in this dataset, but the figures alone do not show the distribution’s shape or explain why they differ. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 4
| U.S. distributed PV segment, 2024 installations | Median size | Average size |
|---|---|---|
| Residential | 7.2 kW | 9.3 kW |
| Non-residential, as defined in the report | 35 kW | 232 kW |
The non-residential figures apply to the report’s defined segment: roof-mounted systems of any size and ground-mounted systems up to 7,000 kW. Within that definition, the median size was 35 kW and the average was 232 kW in 2024. These values should not be read as a measure of all large-scale solar projects; the report’s distributed PV scope has explicit system and mounting boundaries. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 4
State results add context to the residential national figures. California residential systems had a median size of 5.7 kW in 2024. The report says most other states with qualifying observations had medians above 7 kW. State summaries are shown only where at least 20 observations were available, so the phrase “most other states” applies to the states meeting that reporting threshold, not necessarily every state. California also made up a large portion of the residential sample. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 4
Berkeley Lab’s dataset is a project dataset, not a census. The report defines distributed PV to include residential systems, roof-mounted non-residential systems, and ground-mounted non-residential systems up to 7 MWDC. The system-size statistics therefore describe recorded projects inside that scope, not every distributed generation installation in the United States. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update
Distributed generation installations statistics: technology in residential and non-residential systems
Module-level power electronics (MLPE), defined in the report as microinverters and DC optimizers, were used in 87% of residential U.S. systems installed in 2024. Adoption varied by non-residential system segment: MLPE appeared in 73% of small non-residential systems and 29% of large non-residential systems installed that year. These observations show different recorded shares across customer and size segments; they do not establish why those shares differed. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 5
| 2024 U.S. distributed PV segment | Systems using module-level power electronics |
|---|---|
| Residential | 87% |
| Small non-residential | 73% |
| Large non-residential | 29% |
Module efficiency was similar across customer segments in the report’s 2024 observations, with a median of roughly 21%. For residential systems, the stated percentile range was 20.5%–21.5%. The report gives that range specifically for residential systems, so it should not be generalized as a range for every customer segment. Efficiency describes module performance characteristics; it is separate from system capacity, which is expressed in kW in the sizing figures above. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 5
Taken together, the technology observations offer a cross-section of equipment recorded in 2024 installations: MLPE adoption differed by system segment, while the reported module-efficiency median was broadly similar across customer segments. The measurements describe features in the Berkeley Lab project dataset and should be interpreted within its distributed PV scope. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update
Distributed generation installations statistics: storage pairing and battery sizes
Storage attachment measures the share of new PV installations paired with storage; it excludes storage added to existing PV systems and stand-alone storage. In California, residential attachment rose from 14% in 2023 to 57% in 2024. Across all states other than California combined, the residential rate rose from 5% to 8% over the same period. The latter grouping includes Hawaii and is distinct from other report groupings that exclude both Hawaii and California. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 6
| Residential new PV installations | Storage attachment in 2023 | Storage attachment in 2024 |
|---|---|---|
| California | 14% | 57% |
| All states other than California, combined | 5% | 8% |
Hawaii had the highest residential storage attachment rate of any state in 2024, at 85%. The report’s time series says the residential rate in Hawaii remained about 85% from 2021 onward, through the report’s 2024 observation period. This sustained high share is a state-level result for residential PV installations paired with storage, not a national rate. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, pp. 6, 28
Battery capacity varied by paired system segment in 2024. Residential battery storage systems had a median size of 13.5 kWh. Batteries paired with small non-residential PV systems had a median size of 20 kWh, while batteries paired with large non-residential systems had a median size of approximately 550 kWh. The report notes that California represents about 90% of the data used for storage sizing, a relevant limitation when interpreting these values as a picture of the broader U.S. market. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 7
For large non-residential PV-plus-storage systems installed in 2024, at least 20% of paired systems had 4+ hours of storage. This is a duration measure, distinct from battery capacity in kWh: the figures describe different attributes and should not be combined into an inferred output or energy total. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 7
Distributed generation installations statistics: mounting and panel orientation
Ground mounting was common among large non-residential systems throughout the 2010–2024 period. In each year in that span, it accounted for between 35% and 60% of large non-residential systems. Among small non-residential systems, ground mounting accounted for between 13% and 43% per year over 2010–2024. These ranges describe annual shares across the stated period; they are not individual values for a particular year. The mounting statistics include systems with known mounting type. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 26
Tracking equipment was less common in the large non-residential segment: it was used by 1%–9% of those systems per year from 2010 through 2024. That range gives the annual span reported across the period. It does not indicate which year corresponded to either endpoint. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 26
For 2024 installations, 54% of panels faced south and 23% faced west. The report measures primary array orientation, excludes tracking systems, and groups orientations by cardinal directions ±45°. Thus, the orientation percentages describe fixed-orientation arrays classified into directional groups, not every panel including those on tracking systems. Source: Berkeley Lab, U.S. Distributed Solar and Storage: 2025 Data Update, p. 26
Distributed generation installations statistics: reported regional installation totals
Regional installation totals come from different reporting systems and definitions. Minnesota’s Public Utilities Commission reported 34,048 distribution-interconnected DER systems as of December 31, 2025, with nearly 1.86 GW AC of nameplate capacity. The dashboard covers utility-reported systems under 10 MW that operate in parallel with the utility. A utility may be absent if it had not submitted a report, so the total is a reported count rather than a guaranteed census of all qualifying systems in the state. Source: Minnesota Public Utilities Commission, DERs Data Dashboard
Poland’s Energy Regulatory Office (URE) reported 1,636,673 renewable micro-installations connected to the electricity grid at the end of 2025, with almost 13.9 GW of installed capacity. URE defines micro-installations as renewable sources up to 50 kW connected to a grid below 110 kV. During 2025, these installations supplied more than 8.95 TWh of electricity into the grid. The count and capacity are end-of-year figures; the electricity supplied is a total for the year. Source: URE, The number of RES micro-installations in Poland has already exceeded 1.6 million
| Reporting area and measure | Reported observation | Reference period |
|---|---|---|
| Minnesota distribution-interconnected DER systems | 34,048 systems | December 31, 2025 |
| Minnesota DER nameplate capacity | Nearly 1.86 GW AC | December 31, 2025 |
| Poland grid-connected RES micro-installations | 1,636,673 installations | End of 2025 |
| Poland micro-installation capacity | Almost 13.9 GW | End of 2025 |
| Electricity supplied by Polish micro-installations into the grid | More than 8.95 TWh | 2025 |
Poland also reported 272 virtual prosumer installations by the end of 2025. URE said these installations were registered for the first time in 2025; the count was reached during the nearly 6 months after regulations took effect on July 2, 2025. In a separate category, 6 micro-installations operated by collective prosumers had a combined installed capacity of 0.188 MW at year-end. These are specific prosumer arrangements within the Polish reporting context, not additional values that should be added to the national installation total without further category detail. Source: URE, The number of RES micro-installations in Poland has already exceeded 1.6 million
Minnesota and Poland figures should be read side by side as reported totals, not as a direct comparison of deployment rates. Their authorities use different system definitions, thresholds, and reporting coverage: Minnesota reports utility-interconnected DERs under 10 MW, while the Polish figures cover renewable micro-installations up to 50 kW connected below 110 kV. The measures also differ: Minnesota’s cited capacity is AC nameplate capacity, while URE reports installed capacity and annual electricity supplied. Sources: Minnesota Public Utilities Commission, DERs Data Dashboard; URE, The number of RES micro-installations in Poland has already exceeded 1.6 million