Periodic Reporting for period 1 - PVOP (Digitalising the PV sector for the era of Terawatts)
Berichtszeitraum: 2024-05-01 bis 2025-10-31
The world faces a critical situation: climate change driven by greenhouse-gas emissions from massive fossil-fuel use threatens the planet. Energy-related CO2 emissions rose 0.9% in 2022 to over 36.8 Gt, far from the 45% reduction needed to limit warming to 1.5°C. Fossil fuels are also depleting: the oil peak has passed and the gas peak is near. The IEA warns that with current disinvestment, oil production could fall 50% in five years. Meanwhile, the EU remains heavily dependent on external fuel supplies, as shown by recent Russian gas disruptions.
The EU has responded with an ambitious framework—the Green Deal, Climate Law, REPowerEU and Fitfor55—promoting electrification and renewables. Solar PV is expanding rapidly: REPowerEU targets 320 GW of new capacity by 2025 and nearly 600 GW by 2030. SolarPower Europe indicates these goals may be reached early: in 2022 the EU had 209 GW installed and may reach 590 GW by 2026.
Yet this rapid growth brings challenges. PV systems underperform: studies show plants generate 6.3% less than expected. At 1 TW, recovering 75% of these losses would yield 47 GW—more than EU installations from 2018–2020—and €4,072M annually. Causes include undetected failures and the inability of current tools to process huge SCADA datasets. With plants now reaching 300–500 MW or more, engineering teams are overstretched, managing roughly 250 MW per person.
Battery integration could boost performance and flexibility but remains underused due to costs, lack of optimisation tools and regulatory gaps, especially for co-located or centralised storage.
PVOP seeks to address these issues with eight innovative solutions based on models, digital twins and AI combined with expert knowledge. It will apply advanced AI and Big Data to real PV data to uncover causal links behind failures and develop robust solutions. Needs will be validated through Open Science with stakeholders, iterating up to TRL 7. Outputs will include best-practice guidelines and regulatory recommendations, supporting EU goals for climate neutrality and energy autonomy.
PVOP’s overarching goal is to digitalise the PV sector to ensure profitability and meet the 600-GW REPowerEU target by 2030.
Its three specific objectives are:
SO1: Increase PV portfolio performance by 4.7% and cut O&M costs by 32% through near-automatic assessment and fault detection.
SO2: Improve grid-friendly integration, flexibility and energy trading through storage.
SO3: Support EU-wide adoption to ensure PV capacity reliably delivers expected electricity.
The EU has responded with an ambitious framework—the Green Deal, Climate Law, REPowerEU and Fitfor55—promoting electrification and renewables. Solar PV is expanding rapidly: REPowerEU targets 320 GW of new capacity by 2025 and nearly 600 GW by 2030. SolarPower Europe indicates these goals may be reached early: in 2022 the EU had 209 GW installed and may reach 590 GW by 2026.
Yet this rapid growth brings challenges. PV systems underperform: studies show plants generate 6.3% less than expected. At 1 TW, recovering 75% of these losses would yield 47 GW—more than EU installations from 2018–2020—and €4,072M annually. Causes include undetected failures and the inability of current tools to process huge SCADA datasets. With plants now reaching 300–500 MW or more, engineering teams are overstretched, managing roughly 250 MW per person.
Battery integration could boost performance and flexibility but remains underused due to costs, lack of optimisation tools and regulatory gaps, especially for co-located or centralised storage.
PVOP seeks to address these issues with eight innovative solutions based on models, digital twins and AI combined with expert knowledge. It will apply advanced AI and Big Data to real PV data to uncover causal links behind failures and develop robust solutions. Needs will be validated through Open Science with stakeholders, iterating up to TRL 7. Outputs will include best-practice guidelines and regulatory recommendations, supporting EU goals for climate neutrality and energy autonomy.
PVOP’s overarching goal is to digitalise the PV sector to ensure profitability and meet the 600-GW REPowerEU target by 2030.
Its three specific objectives are:
SO1: Increase PV portfolio performance by 4.7% and cut O&M costs by 32% through near-automatic assessment and fault detection.
SO2: Improve grid-friendly integration, flexibility and energy trading through storage.
SO3: Support EU-wide adoption to ensure PV capacity reliably delivers expected electricity.
The work carried out follows a clear philosophy: first, to ensure that operational data monitored in PV plants is of sufficient quality for use by artificial intelligence algorithms; and second, to develop AI and big data solutions for automatic fault detection and diagnosis, thereby increasing plant performance and profitability.
In the first project period, activities focused on two main lines:
• Developing concepts for PV plant sensor solutions, AI-based fault-detection tools and storage-management strategies. Key Performance Indicators (KPIs) were also proposed to assess whether these solutions function properly.
• Based on these concepts, alpha versions of the solutions were created—laboratory prototypes validated under controlled conditions.
Validation tests quantified the KPIs and delivered very satisfactory results, allowing the project to enter its second phase with solid expectations. The next step is the development of beta versions of the technical solutions, which will be validated in real PV plants under real operating conditions.
In the first project period, activities focused on two main lines:
• Developing concepts for PV plant sensor solutions, AI-based fault-detection tools and storage-management strategies. Key Performance Indicators (KPIs) were also proposed to assess whether these solutions function properly.
• Based on these concepts, alpha versions of the solutions were created—laboratory prototypes validated under controlled conditions.
Validation tests quantified the KPIs and delivered very satisfactory results, allowing the project to enter its second phase with solid expectations. The next step is the development of beta versions of the technical solutions, which will be validated in real PV plants under real operating conditions.
The development of PVOP activities will lead to the following results:
- R1: Sensorisation toolkit to measure the operating variables with low uncertainty.
- R2: Free-access tool for the simulation of the PV plants productivity with digital twins adjusted with sensor data.
- R3: AI-based Electricity Market Prediction Tool.
- R4: Smart tracking control for large mono and bifacial single-axis tracking PV plants over terrain of arbitrary orientation and slope taking into account diffuse irradiance conditions and extreme meteorological events.
- R5: AI-based smart control system of PV plants with batteries with weather and electricity market forecasts to maximise the energy trading.
- R6: Near 100% automatic AI-based system based on a family of algorithm for faults detection and diagnosis and optimisation of resources in PV plants.
- R7: Near 100% autonomous aerial inspection and fault detection system based on the integration of AI and high temporal and spatial resolution multispectral aerial imaging solutions.
- R8: Near 100% automatic and predictive PV asset management software to maximise performance and optimisation of asset O&M management based on multi-level KPIs and integrating data from sensors, images and reports from O&M teams.
- R9: Policy recommendations to improve design practices, policies and decision making, and to raise consumer awareness.
- R10: Guide of best practices in the design of PV plants for good sensorisation and implementation of AI-based systems for automated operation.
- R11: Public and wide database of operational data of PV plants to enable, foster and empower AI for Digital PV at European scale.
It is estimated that these results will have the following impact on the PV sector and on European society:
- I1 - Increase in performance: 4.7%.
- I2 - Increase in annual profitability: €4M/GWp.
- I3 - O&M and PV asset management cost reduction: 32%.
- I4 - Reduction in the number of false alarms in the management system: 90%.
- I5 - Reduction of the time spent on operation evaluation: 80%.
- I6 - PV system reliability: Reliability extended by automatic failure detection procedures and management of maintenance teams.
- I7- PV system security and flexibility: Integration and management of batteries increases flexibility and security of electrical grid.
- I8- Enhanced digitalisation: AI-based solutions increase digitalisation of PV systems.
- I9 - ROI of batteries in PV systems: 5-8 years (depending on external factors).
- I10 - Utility-friendly integration into energy system: Integration and management of batteries increases the utility-friendly integration by strategies like AS, DM or AGR.
- R1: Sensorisation toolkit to measure the operating variables with low uncertainty.
- R2: Free-access tool for the simulation of the PV plants productivity with digital twins adjusted with sensor data.
- R3: AI-based Electricity Market Prediction Tool.
- R4: Smart tracking control for large mono and bifacial single-axis tracking PV plants over terrain of arbitrary orientation and slope taking into account diffuse irradiance conditions and extreme meteorological events.
- R5: AI-based smart control system of PV plants with batteries with weather and electricity market forecasts to maximise the energy trading.
- R6: Near 100% automatic AI-based system based on a family of algorithm for faults detection and diagnosis and optimisation of resources in PV plants.
- R7: Near 100% autonomous aerial inspection and fault detection system based on the integration of AI and high temporal and spatial resolution multispectral aerial imaging solutions.
- R8: Near 100% automatic and predictive PV asset management software to maximise performance and optimisation of asset O&M management based on multi-level KPIs and integrating data from sensors, images and reports from O&M teams.
- R9: Policy recommendations to improve design practices, policies and decision making, and to raise consumer awareness.
- R10: Guide of best practices in the design of PV plants for good sensorisation and implementation of AI-based systems for automated operation.
- R11: Public and wide database of operational data of PV plants to enable, foster and empower AI for Digital PV at European scale.
It is estimated that these results will have the following impact on the PV sector and on European society:
- I1 - Increase in performance: 4.7%.
- I2 - Increase in annual profitability: €4M/GWp.
- I3 - O&M and PV asset management cost reduction: 32%.
- I4 - Reduction in the number of false alarms in the management system: 90%.
- I5 - Reduction of the time spent on operation evaluation: 80%.
- I6 - PV system reliability: Reliability extended by automatic failure detection procedures and management of maintenance teams.
- I7- PV system security and flexibility: Integration and management of batteries increases flexibility and security of electrical grid.
- I8- Enhanced digitalisation: AI-based solutions increase digitalisation of PV systems.
- I9 - ROI of batteries in PV systems: 5-8 years (depending on external factors).
- I10 - Utility-friendly integration into energy system: Integration and management of batteries increases the utility-friendly integration by strategies like AS, DM or AGR.