The operations and maintenance (O&M) of solar plants is a key market area where technologies are continuously evolving. With the rapidly increasing size and scale of solar projects, there is now a greater focus on optimising performance and reducing operational costs. The rising digitalisation and automation of solar photovoltaic (PV) plants has helped increase the availability of performance-related data. Using cutting-edge digital tools, the data can be analysed to add value and reduce the O&M costs of solar plants. A key application of digital technology is predictive maintenance.
The application of analytical monitoring systems can help in detecting possible malfunctions by assessing the system performance of a PV unit. Several external factors can impact the actual performance of the system. It is, therefore, important to ensure that failures and unanticipated degradation issues do not go undetected. Any lapse in detection may lead to a loss of energy generation. Predictive maintenance refers to actions performed to assess the condition of a solar PV system and its components, and recommend when maintenance should be undertaken. The prediction is derived from the analysis and evaluation of the data collected from the system, including significant parameters such as those related to degradation. Based on a cost-benefit analysis, monitoring systems and personnel expertise can help decide the appropriate course of action.
For effective O&M, data should be properly collected and analysed to identify anomalies. Digitalisation and automation tools play a key role in increasing the efficiency of this process given the large volumes of data. Automatic data acquisition is integrated with advanced monitoring and analysis systems based on artificial intelligence (AI) and machine learning software to optimise and enhance system performance. It is important to store all the data and workflows to create automatic logbooks of O&M and alarms. Future performance can be improved from the lessons learnt in the past as well as ongoing O&M practices.
Moving to predictive
As per the European standard EN 13306, predictive maintenance is defined as “condition-based maintenance carried out following a forecast derived from repeated analysis or known characteristics and evaluation of the significant parameters of the degradation of the item”. There is a growing focus on the prediction of likely anomalies and identification of potential causes to take pre-emptive action. This is done by regular monitoring and analysis at the component level of the DC array, transformers, inverters, combiner boxes and strings.
Anomalies are identified before the next circuit testing or thermal imaging inspection. For a solar plant, maintenance is usually carried out on-site by specialised technicians or subcontractors, in close coordination with the operations team. Among the services offered by O&M contractors, predictive maintenance is a special service. While supervisory control and data acquisition (SCADA) systems are popular for remote monitoring and management, more sophisticated technologies are now being adopted to address the limitations of SCADA. Over the past few years, various technologies have emerged to improve predictive maintenance. These technologies have evolved from computer-based automated monitoring systems to drones, robots and wearables, and now to solutions based on AI and internet of things (IoT).
Advanced technologies are being deployed to improve predictive maintenance. One such technology is the software platform called EIRA, developed by Inspire Clean Energy, a Mumbai-based O&M service company. EIRA is based on the three-seconds and three-clicks concept, whereby a client is made aware of the status of a solar plant in three clicks. EIRA logs into every project at 15-minute intervals to check the performance parameters of each inverter, energy meter and transformer. If a flaw is detected, it automatically generates a ticket and sends it to the nearest engineer. As soon as the problem is solved, a report is sent to the client.
O&M can be made more effective by adopting best practices to improve the health of the system. On the supply side, manufacturers can provide a list of status and error codes produced by the device. These status and error codes must be standardised for all manufacturers. Further, solar system equipment should be equipped with sufficient sensors and opt for an appropriate monitoring software system, which can provide basic trending and performance comparison between time intervals as well as sites. Many operators have now started moving towards intelligent predictive maintenance based on advanced sensors and IoT.
Best practices for the operations team include implementation and development of procedures to effectively analyse historical data and identify behavioural changes that might jeopardise system performance. Behavioural changes are usually related to predetermined or unpredicted equipment degradation processes. The maintenance team can implement predictive maintenance to prevent any possible failures, which can cause safety issues and lead to energy generation losses.
Optimised hardware replacement
The levellised cost of electricity of solar power is heavily affected by the operational cost of solar plants. To maintain the economic viability of a plant, it is critical to optimise its operational quality, reduce maintenance costs and maximise performance. One of the ways to achieve this is working towards minimising the number of times the parts are replaced during the project life owing to wear and tear. In addition to preventive maintenance, predictive maintenance can be undertaken when there is unexpected deviation in performance.
To find the optimal balance between the costs and benefits of maintenance interventions, different maintenance optimisation models are used based on the probabilities of component failure. By modelling the uncertainty in the time-to-failure with a known probability distribution function, an appropriate predictive monitoring system could help in assessing the optimal hardware replacement cycle. Big data analysis is also used for optimisation. It enables the easy recognition of faults, provides a clear diagnosis in some cases and recommends short-term actions to avoid probable upcoming issues. While this technique of predictive maintenance can lower the cost of ineffective scheduled maintenance and reduce device downtime, the methods are usually sensitive to device models and brands, and restrict generalisation.
Solar power technology is evolving from computer-based systems, databases and communication networks and servers to IoT- and cloud-based platforms, advanced analytics, predictive data analytics, asset performance management software, smart sensors and intelligent forecasting solutions. With rapid technology advancements in automation and digitalisation, and resultant improvements in O&M efficiency and cost, it is expected that the majority of O&M activities will be carried out through digitally enabled techniques in the near future. The rise in digitalisation can help in improving plant efficiencies on a real-time basis, leading to enhanced equipment life. Digital systems can perform multiple functions in the solar O&M space such as improvement in power plant efficiency, reduction of operational expenses, and mitigation of unplanned outages.
Predictive maintenance can help minimise losses in energy production, and enable better decision-making and optimal utilisation of overall cost-saving techniques. The economic benefit is all the more attractive, given the falling tariffs for solar projects and the advent of the Covid-19 pandemic. O&M players are being pressured to run solar power plants efficiently at low costs while maintaining profitability. Further, moving to technologically advanced solutions would require less involvement of on-site personnel activities, which is desirable in situations like the pandemic.
Positive results from predictive maintenance have led an increasing number of players towards the adoption of digitalised solutions in O&M, along with its promise of profitability. Moving forward, remote monitoring, automation and predictive maintenance will be focused on minimising costs and maximising efficiency.
By Meghaa Gangahar