Solar forecasting · Earth observation · Machine learning
Quentin Paletta
Co-founder & Chief Information Officer at Wematics — building AI-driven sky cameras for real-time solar forecasting. 📈
Previously a Research Fellow & Technical Officer at ESA's Philab, with a secondment at the International Energy Agency, following a PhD at the University of Cambridge on deep learning for solar nowcasting.
About
From cloud physics to grid-scale forecasting
I'm interested in computer vision, machine learning, Earth observation and energy — and in turning that research into tools people actually use to run the grid.
At Wematics, the company I co-founded, we build a next-generation AI-driven sky camera that offers real-time cloud tracking and solar irradiance forecasting, helping grid operators and energy providers integrate more renewables and reduce costs.
Before that, I was a Research Fellow and Technical Officer at the Philab of the European Space Research Institute in Frascati, in collaboration with the Climate Office of the European Space Agency in Harwell — working on Earth observation and machine learning for climate and energy, including a secondment at the International Energy Agency exploring the potential of Earth observation for the energy sector. And before that, at the University of Cambridge, I developed the deep learning algorithms for solar nowcasting that Wematics is now built on.
Also working on solar energy meteorology? I run a Slack community of 130+ researchers and engineers from around the world — don't hesitate to reach out to join.
Experience
Timeline
-
Co-founder & Chief Information Officer CurrentWematics
Developing an AI-driven next-generation sky camera offering real-time cloud tracking and solar irradiance forecasting — empowering grid operators and energy providers to optimise renewable energy integration and reduce costs.
-
Research Fellow & Technical OfficerPhilab, European Space Research Institute (ESA) · Frascati
Earth observation and machine learning for climate and energy, in collaboration with the ESA Climate Office in Harwell — including a secondment at the International Energy Agency exploring the potential of Earth observation for the energy sector.
-
PhD, Engineering DepartmentUniversity of Cambridge
Developed deep learning algorithms for advancing solar nowcasting based on cloud cover observations from sky cameras and weather satellites. Thesis: Vision-Based Solar Forecasting with Deep Learning.
Current venture
Wematics — AI-driven sky cameras for solar forecasting
As Co-founder and Chief Information Officer at Wematics, I'm committed to advancing the use of sky cameras for solar applications. Our next-generation sky camera pairs high-resolution cloud imagery with deep learning to give grid operators and energy providers a live, minutes-to-hours-ahead view of incoming solar power.
- → Real-time cloud tracking from ground-based sky cameras
- → Deterministic and probabilistic solar irradiance forecasting
- → Built for grid operators and energy providers integrating more renewables
How it works
-
1
CaptureThe sky camera continuously images local cloud cover.
-
2
PredictAI models track cloud motion and forecast solar irradiance from seconds to hours ahead.
-
3
IntegrateOperators use the forecasts to optimise renewable integration and cut balancing costs.
Research
Selected publications
Much of my research develops methods that use cloud cover observations — from sky cameras and satellites — for solar energy forecasting. A selection is below.
A follow-up to the IEA's landmark Energy and AI report, examining how the energy–AI nexus has evolved amid surging data centre investment.
My PhD thesis — the technical foundation Wematics is now built on.
A multilocation satellite and sky image dataset for solar forecasting and atmospheric sciences.
Zero-shot and few-shot learning tasks based on a model pre-trained at another location.
Literature review on computer vision-based solar forecasting with deep learning applied to sky cameras and meteorological satellites.
Combining sky images and satellite observations into a single ML framework for intra-hour solar forecasting.
Community
Let's talk solar energy meteorology
Looking for help to find or conduct a project on Earth observation, climate and/or energy — or just working in solar energy meteorology? I run a Slack community of 130+ researchers and engineers from around the world. Don't hesitate to reach out.