Predictive Maintenance of Photovoltaic Energy Generation Plants – DILIGENT
Company/Organisation: Nexlys Lda
Project Description
Challenge
The integration of renewables into power systems has led to multiple studies and analysis in terms of grid-power quality, reliability, and/or feasibility. Solar energy is the radiant energy emitted from the sun, which is harnessed by using various technologies such as solar heating, photovoltaic cells, and others. It is an efficient form of unconventional energy and a convenient renewable solution towards growing greenhouse emissions and global warming. With regard to photovoltaic (PV) installations, monitoring problems requires detailed analysis, since solar-radiation fluctuations, soiling on solar panels, or deficiency of PV-panel performance can involve unexpected power-output oscillations and, subsequently, undesirable power-generation oscillations.
Proposed solution
The Predictive Maintenance of Photovoltaic Energy Generation Plants (DILIGENT) is a data-driven solution focused in the delivery of machine learning based recommendations towards the photovoltaic energy generation plant production, safety and maintenance staff. It leverages in technologies like sensors, satellite data, meteorological data, IoT and AI.
The DILIGENT’s AI algorithm makes use of a combination of supervised machine learning and deep learning (TensorFlow based) technologies. Aside of the PV onsite historical archive, the DILIGENT AI Server also collects data from the from the Copernicus Atmosphere Monitoring Service (CAMS).
Nexlys Lda. has developed a prototype, with the collaboration of a photovoltaic energy producer.
During August 2022, the technology was validated in a laboratory environment and while using historical energy production data, maintenance records, IoT environmental sensor data and CAMS data, and hence achieving the Technology Readiness Level (TRL) 4 – technology validated in lab.
At the end of the proposed activities the DILIGENT technology will achieve TRL 6 – "technology demonstrated in relevant environment", while being demonstrated at the Fotovoltaica Macotera SL photovoltaic energy generation plant.
