In a world of rising capital costs and the associated need to maximize project returns, technology is becoming increasingly important. Thanks to AI and machine learning, the vast amounts of data that solar farms produce can be processed, analyzed and built into powerful strategic insight.
That was the gist of the title of the session How is technology driving UK solar energy forward? on day two of British sun top, where moderator Abid Kazim, founder of Lumina Renewables, opened the session by stating: “The cost of capital has become an issue. We can’t throw money at solar power stations anymore, we have to make them work better.”
With that in mind, the industry is turning to technology to provide new approaches to greater solar energy efficiency, changing the way we think about and measure success.
Less is more
The conventional approach to project design has been to orient solar panels toward the south to maximize energy production. To produce more power and therefore greater efficiency, simply add modules. However, rising capital costs, both for financing additional modules and for the land to install them, mean that this approach no longer delivers the returns it once did.
Eric Montméan, head of UK solar development at TotalEnergies, spoke about a French solar farm where the modules are oriented westwards. This means that, unlike south-facing solar, the modules capture more energy later in the afternoon, when market prices are higher, and provide a higher ROI, despite producing fewer watts overall. The result? A 16% increase in turnover by using more electricity at the end of the day. TotalEnergies is currently testing this methodology in Britain and the signs are promising. Skewed generation can generate more revenue per watt even if you generate fewer watts.
Limit losses
Artificial intelligence in general, and machine learning in particular, can not only revolutionize the process of planning new projects and delivering greater returns on existing projects, but can also be used to identify problems with existing installations .
Emilio Martinez, UK head of technical consultancy and asset management at Vector Renewables, praises the benefits of using artificial intelligence to identify problems with existing projects.
“We had a customer who invested in a brand new 50 MW PV installation with state-of-the-art trackers. During the first and second years, production was approximately 6% below expectations. For such a large project, 6% has a huge impact on the economics of the project,” said Martinez.
“We used machine learning to replicate the real PV installation according to mathematical models. We used different models over a few months. You have 200 inverters, add the trackers and the amount of data is enormous. Our models predicted production with an accuracy of 0.4%.”
The EPC engineer stated that there were no easily identifiable issues, and they were right, but the team’s modeling found the trackers to be operating 4% below spec. Furthermore, the modules themselves did not function according to the expected characteristics: they were 2% below the claimed output.
These numbers can be attributed to temperature issues, with silicon performing less effectively at higher temperatures, but that wasn’t the problem. Laboratory testing revealed that the modules did not operate according to specifications at the stated output. However, the manufacturer did not accept that the modules were defective.
While the insight did not result in financial compensation for the project owner, it did lead to Vector’s customer including clauses in future contracts for panel performance and tracker profits to mitigate future issues.
The new technological frontier
Machine learning allows us to measure everything and process that information at scale and speed. Tens of thousands, hundreds of thousands or millions of data points can be processed in an instant, allowing problems to be identified almost immediately. The sheer amount of data that can be ingested and analyzed is staggering – far more than can be achieved manually.
By using this technology to assess and reassess projects, efficiencies can be found in the planning, construction and operations processes. That means costs can be kept under tight control, as can revenue, which could mean the difference between a project getting the green light or being cancelled.
Concluding the session, Kazim summarized this new technological frontier: “The biggest challenge we face now, and will face in the near future, is Europe’s need for capital to build new power plants. That requires every cent we have available, and if you can’t make the money, you can’t build more factories.
“So we have to find creative ways to maximize value. Technology is changing the conversation.”