AI-driven solar panel assessments
Feb 2025
See how Nearmap AI-driven solar panel assessments streamline nationwide installation planning for scalable, cost-effective renewable energy solutions.
Feb 2025
As solar energy adoption continues to rise, organisations supporting green initiatives face the challenge of performing scalable solar panel assessments. Manual field inspections are time-intensive and costly, making large-scale assessments nearly impossible to execute quickly.
Using location intelligence to assess solar panel installations across entire regions saves time and is more cost efficient ways. A recent collaboration between Nearmap and a green loan provider demonstrates how.
The challenge: Nationwide solar panel identification
One of our customers, a provider of green loans for solar installations, wanted to identify consumers who already had solar panels installed so it could offer them reduced-cost funding for environmentally friendly appliances.
Conducting nationwide field inspections to locate these solar installations would have been prohibitively expensive and time-consuming. So, the business sought accurate, reliable solar panel installation data at a national scale that would replace the need for sending inspectors to every address.
Demonstrating AI accuracy in solar panel identification: A proof of concept
To test the accuracy of Nearmap AI solar panel data, the customer selected 19,000 random test sites nationwide to verify whether our AI-driven data could detect existing solar panels.
Using a QGIS algorithm to randomly select a 2% sample — about 377 addresses — for closer inspection, we put our AI data and high-resolution imagery to the test. The objective was to ensure that Nearmap could deliver a high degree of accuracy when identifying solar panels. If successful, it would allow the customer to proceed confidently with reduced-cost funding for green loans.
The data review process: Identifying true positives and false negatives
The Nearmap team conducted a thorough visual analysis of the randomly selected addresses, using a combination of AI-detected features and human review to ensure accuracy. Key points of discussion with the customer centred around several important topics:
- False negatives: Instances where solar panels were present but not detected by AI.
- Duplicate addresses: Multiple addresses linked to a single building, leading to potential confusion in solar panel identification.
- Null addresses: Empty parcels with no buildings, which were mistakenly flagged as potential addresses.
These findings were important not only for transparency but to build trust. By sharing the limitations and strengths of the AI dataset, we ensured that our customer fully understood how the data was processed and interpreted.
One of the main challenges was the issue of duplicate addresses. For example, in a block of apartments where only one unit had solar panels, multiple addresses were associated with the same building. This type of data inconsistency could lead to incorrect conclusions, so the team carefully explained these nuances to the customer.
An example of how Nearmap AI identifies solar panels (and swimming pools, in this example) is shown below, using a random site selected in Butler, Western Australia.
Butler, WA AU
Nearmap AI highlighting solar panels (yellow) and swimming pools (blue)
Accuracy results: 98.6% success rate
After assessing the 377 buildings, the results were promising:
- 27 addresses were true positives, meaning the AI correctly identified the presence of solar panels.
- 345 addresses were true negatives, confirming that no solar panels were present, as the AI correctly indicated.
- 4 addresses were false negatives, where solar panels were present but not detected by the AI.
Using AI for renewable energy initiatives, we were able to achieve a 98.6% accuracy rate in identifying solar panel installations, which confirmed the high reliability of the Nearmap solar dataset and gave the business confidence to pursue the solar panel identification process, using Nearmap data as the foundation for offering reduced-cost funding to existing solar users.
)
)
Driving smarter green loan decisions with AI
The success of this project highlights the value of AI in streamlining and scaling solar panel assessments. What would have taken weeks or months of manual field inspections was accomplished in a fraction of the time.
Nearmap AI-derived solar panel data provided significant savings for this green loan solar energy provider, enabling it to offer targeted funding for environmentally friendly appliances.
The future of scalable solar assessments
As demand for renewable energy increases, the need for precise and scalable data to monitor solar installations across vast areas becomes critical.
Through AI-powered insights and high-resolution aerial imagery, Nearmap is at the forefront of this shift. Our tools equip utility providers and green loan companies with the power to conduct large-scale solar panel assessments with the data required to make informed decisions that drive a sustainable future.
Get in touch today to discover how we can help you power your renewable energy initiatives.