As the installed capacity of solar PV continues to expand, the requirements for technical O&M are growing too. The industry has been driven by significant cost efficiencies over the years, both in Capex (Installation) and Opex (O&M) but the big gains have already been made in these areas and future gains are likely to be marginal. As a result of expanding capacity and the need to continue reducing overhead, digitalisation has become a key theme in the solar industry.
Digitalisation of O&M was a major topic at the PV Operations Europe 2019 conference in Munich, continuing from last year’s discussions. Big data is likely to become a significant influence in managing the multitude of data streams associated with a typical solar project – SCADA, spare parts information, inspection reports from different providers such as technical advisories, and predictive tools for predictive maintenance.
Data systems can be used for anything from asset performance (monitoring the standard KPIs, typically from performance ratios), plant availability and how it performs throughout the seasons, to security systems.
This brings a number of questions to resolve around data aggregation, monitoring, analysis, remote versus centralised data, cloud storage, and whether a company should manage their data-driven operations in-house, or outsource to third-party O&M monitoring providers.
Of course, the correct solution for a better performing plant is one that works for each company, depending on a combination of factors, like location and the size of the asset portfolio being managed. It’s where operating and maintenance come into play together with asset management.
Who manages the data?
A typical investment fund or an asset manager has numerous streams from various monitoring systems that they use to collect and analyse that data. Now the industry aims to aggregate these streams into a single unified software that makes it easy to manage the data.
Much like the theme at last year’s conference around managing O&M in-house or outsourcing the work, the question for data management is whether you should outsource your monitoring solutions to an asset manager or a monitoring solutions company like Meteocontrol or 3Megawatt, or whether to bring it in-house with your own technical asset management team.
There are a number of contract companies who offer data monitoring solutions. For example, one of our clients near Munich currently monitors 13 Gigawatts of assets across forty-five thousand projects. On the other hand, a lot of manufacturers have their own monitoring system in place, and there is also a standard SCADA system.
One advantage of using a service provider is their access to a larger dataset across multiple assets, which may lead to greater efficiency and better decision-making when it comes to choosing how to operate an asset. On the other side of that, in-house data analysis might bring a deeper understanding to the technical asset manager for their particular asset, based on location etc.
Another company that presented in Munich has a central operating system purely because the majority of their sites are particularly isolated and require processes and platforms to operate their remote plants.
There is only so much a technical asset management team can deal with in terms of workload. Some companies have been able to implement big data effectively and the choice to outsource depends on the software and the budgets they have in place, how they operate and maintain their portfolio and how they analyse the data.
How can we use Big Data in Solar PV?
Data can help us predict future performance, whether it’s energy availability or the efficiency of the asset. In some cases, where extreme weather conditions have damaged some of the plants, or you need a turn off a plant for preventative maintenance, there can be a detrimental effect on the financial return which is one of the most important factors an investor looks at.
There is a lot of potential for forecasting, similar to the way Deepmind can use weather forecasts to boost wind energy demand. There’s a lot that solar can learn from wind, which typically experiences more faults than solar. Wind producers have had to learn how to overcome these faults or prevent them from happening and wind O&M practices are at a more mature stage than for solar.
With the ability to connect and monitor solar, whether we build large solar plants or localised rooftop installations depends on the regulatory environment. In countries without subsidies, it would be difficult to build a large-scale project on the ground whereas it would be much easier to obtain sign off for commercial, industrial, and rooftop projects. Besides, rooftop projects typically provide power only for that particular building, which makes approval easier.
For a large-scale project in a greenfield site or a remote area, the developer needs to consider the end user and might require a PPA. Commercial and industrial projects seem to be quite popular at the moment, especially in Southern Europe. Portugal is building a lot of rooftop projects. But in countries like Spain and Italy, large utility-scale ground projects are popular. Some of the projects in Spain are in the one hundred megawatts region. The largest in the UK is around sixty or seventy megawatts. Older markets are coming back on board which is encouraging for the industry.
A lot of companies will be trying to standardise and centralise data streams. So far, it’s mainly the leaders with large portfolios – investment funds or IPPs (independent power producers) who have adopted big data. These organisations typically have assets all over the world – in Europe, Asia and Africa – and it’s much harder to manage the data streams without a centralised system.
In terms of employment opportunities, big data is also bringing new categories of jobs to solar, like Software Developers, and Data Analysts with a strong mathematical background, especially for monitoring services companies.
It’s important that we focus on what’s happening now, rather than looking only to the future, which is likely to focus on the trend of solar companies using big data. But right now, we can focus on perfecting what we’re currently doing – perfecting performance ratios, availability, financial returns, O&M and HSE, to meet the base objective and the numbers required to get the financial returns from each asset.
The economics of solar has been a vital driver of the industry and solar O&M is becoming cheaper. Jenny Chase of BNEF highlighted the 73 per cent drop in the operating cost of solar from 35,000 EUR/MW/year in 2011 to 9,500 EUR/MW/year in 2018. It’s becoming cheaper to operate the plants, which makes room for more financial returns and draws further investment in the technology.