- CalibSun | Expertise
Experts in
Solar Energy
18+ years experience of PV engineering for solar players
Making them work coherently across scales, regions, and sources is the real engineering challenge.
In an industry where global renewable energy capacity is growing at record pace and grid management is under mounting demand pressure, the ability to access, process, and act on solar resource data in real time is a market requirement.
At CalibSun, this is precisely the challenge we’ve been solving thanks to our photovoltaic engineering inherited from Solaïs, a pioneering PV engineering consultancy founded in 2008. Our solutions are backed by robust academic research, developed in partnership with a key contributor to the International Energy Agency (IEA) specializing in solar irradiance, meteorology, and data. Driven by a deep commitment to R&D, we actively advance the science of solar energy through ongoing research, peer-reviewed publications, and contributions to international conferences.
Our Solar expertise
Quality Check of sensor data
On a PV plant, data integrity directly conditions every downstream process: solar and power forecasting, performance monitoring and asset management, and solar resource assessment. CalibSun applies a two-phase quality check: a one-time historical data QC pass at onboarding for machine learning, and a continuous automated flagging process on every incoming measurement.
GHI spatialisation
CalibSun is the only provider that natively integrates on-site data and performs true GHI spatialization. The method is applied operationally on large-scale plants, and ranked first of seven forecast providers in a three-month blind benchmark run by the client across thirty PV sites. This reflects the true GHI of the plant far more accurately than any independent sensor. It produces a significantly stronger correlation with measured power production.
On-site data integration
CalibSun integrates on-site solar data collected directly at a photovoltaic power plant to monitor real weather conditions in real-time. For solar power forecasting, these on-site measurements are the primary source of truth that satellite imagery and numerical weather prediction (NWP) models cannot replace at short forecast horizons.
Weather data
Weather data for solar forecasting combines 3 families: NWP, satellite imagery, ground observations. CalibSun’s meteorology team integrates the full set of meteorological variables used to estimate solar irradiance and anticipate PV production.
Solar science: Three layers, one forecast
These three layers do not serve the same horizon. Weather models carry the day-ahead, satellite imagery the intraday, on-site measurement the shortest horizons. For long-term resource assessment the relationship reverses: the on-site campaign calibrates the satellite history, and it is that corrected history, not the campaign itself, that sizes the project.
On-site data
Pyranometers measuring GHI, GTI and meteorological sensors, and SCADA telemetry capturing power output.
Satellite images
Spatial coverage at regional or continental scale from geostationary satellites at high temporal frequency
Weather models (NWP)
Solar irradiance as a by-product of full atmospheric state prediction at horizons from a few hours to several days ahead.
- CalibSun’s resources
Our resources
Customized for you
Research papers, technical notes and press coverage from the CalibSun team. Our publications document how in-situ measurement and sky imaging improve solar forecast accuracy, and how producers and traders turn that accuracy into revenue.
- Questions
Solar expertise: your questions answered
What is the difference between solar data and weather data?
- Weather data encompasses the full atmospheric state, wind speed, precipitation, temperature, pressure, humidity, storm patterns, cloud cover, as monitored by networks of ground stations and reported by government agencies including NOAA and national meteorological services worldwide.
- Solar data is a focused subset, centred on radiation variables: GHI, DNI, DHI, and their derived quantities. In practice, solar forecasting pipelines combine both: weather data provides the atmospheric context, while variables including solar radiation at the surface are the direct inputs to energy yield models.
Can solar data be acquired without on-site sensors?
Yes, satellite-derived products and NWP outputs allow solar resource estimation at any location on Earth without a single ground sensor. Public databases give access to solar data across virtually every region and country, and are widely used at the early stages of a solar project for resource screening and installed capacity estimation. However, the absence of on-site data carries a precision cost: integration of real-time irradiance and on-site power production data enables forecasts to be refined, improving the accuracy of CalibSun’s NEXT by 46% at 5 minutes and 33% at 15 minutes against the incumbent forecasting provider. On-site data is not mandatory; it is a performance multiplier.
What happens to forecast accuracy under difficult climatic conditions?
Standard NWP-based term forecasts degrade significantly in tropical, island, high-altitude, or monsoon-affected climates, where cloud dynamics are localised and highly variable, and where installed solar capacity is growing fastest. Conventional models are not sufficient in these contexts: weather conditions are too localised and changeable, and real-time sky monitoring becomes essential.
CalibSun’s INSTANT solution anticipates variations in photovoltaic production by detecting clouds up to 30 km away using a network of fisheye cameras, enabling utility operators and power plant managers to adjust thermal generation up to 30 minutes before solar output falls.
