Predictive Maintenance project

Improving performance of PV plants by predicting failures

Predictive Mainteance, man with a laptop and solar panels

In the search for renewable energy sources the growth of solar energy plants has been remarkable. DNV’s Predictive Maintenance for Solar PV Plants project aims to improve solar plant’s efficiency by using predictive maintenance techniques.

Contact us:

Gerardo Guerra

Gerardo Guerra

Senior Researcher

Pau Mercade

Pau Mercade

Senior Researcher

In common with most industries, maintenance of solar photovoltaic (PV) plants is generally done using preventative or corrective methods. This means inspections and replacement of components at regular intervals or in response to a sudden fault. Most problems are caused by inverter failures.   

Predictive maintenance solution  

DNV and GreenPowerMonitor, a DNV company, have developed a predictive maintenance system for solar inverters that uses machine learning models to represent an inverter’s normal operation and to identify anomalous behaviour within new streaming data. The developed system has been successfully tested. However, to work properly it needs detailed maintenance logs which the majority of systems lack. This has prevented further development and validation of the method.  

As a solution to the lack of detailed logs, DNV and GreenPowerMonitor have explored a more general anomaly detection approach which can be exploited when no maintenance logs for validation of the predictive maintenance system are available. The anomaly detection system relies on the concepts of density and reconstruction error to identify anomalies, and it is complemented with the use of explainability techniques to determine the potential causes of anomalies. The system has been tested using historical data for multiple inverters and results show that the implemented approach can identify sustained anomalies. Anomaly detection can also play an important role as a support tool for a predictive maintenance system but should not be seen as a substitute.

Contact us:

Gerardo Guerra

Gerardo Guerra

Senior Researcher

Pau Mercade

Pau Mercade

Senior Researcher

The stream of operational data is rapidly increasing with each new solar project being added to customer portfolios. However, operational data does not itself deliver value, customers need actionable insights to help them operate and maintain their expanding solar fleets. The predictive maintenance capabilities GRD is developing in partnership with GPM and Energy Systems, will activate solar operational data to allow asset manager and operators to anticipate O&M needs and maintain system performance.
Dana Olson,
  • Solar Segment Leader
  • DNV

The benefits  

Predictive maintenance can help to anticipate failures and optimize maintenance interventions, which will result in an increase in the plant’s availability and performance. Anticipating failures will allow O&M service providers to better manage and plan staff assignments, spare parts, and replacements. Decreasing the number of unexpected failures reduces the cost of electricity as it eliminates the need for back up reserves and it also contributes to a more stable grid.  

Developing a standardized system of predictive maintenance now will allow these benefits, and the deployment of anomaly detection systems can help to close the gap between the current situation and the next generation of predictive maintenance systems. 

The studies conducted so far have demonstrated the efficacy of current techniques and the general solution framework which could be applied to the different components within the PV plant, but more work needs to be done before predictive maintenance becomes a ubiquitous tool across the industry. 

Although commercially available predictive maintenance systems are being developed by manufacturers, system operators have requested independent solutions that can be uniformly deployed across their technologically diverse system portfolios. Third-party independently developed systems can provide an unbiased, more robust approach that can complement and verify the findings from proprietary systems. It is important, however, for all industry stakeholders to act and address the issues preventing predictive maintenance from reaching its full potential.