When it comes to renewable power sources such as solar, technological innovations have made great strides in improving the collection and distribution of energy.
From more efficient photovoltaic cells to battery storage methods, these advances are helping to make solar a genuine player in the energy market.
Engineers, however, are doing more than looking at the cells themselves and considering what happens in the many years of operation after they are installed.
Presenting at Engineers Australia’s Climate Smart Engineering conference, Ido Molad, Vice President of Research and Development at Israeli company Ecoppia, turned his attention to a significant maintenance problem for utility-scale photovoltaic sites: soiling.
“The loss to the power generation that can be caused by soiling, in some areas in the world, can reach up to 50 per cent,” he told the conference in his presentation, “Robotics development challenges for a rapidly growing market”.
“The soiling has not just daily effects on the power generation, it can also cause degradation of the panel’s performance in time due to that soil. So clearly we understand that cleaning is essential on a daily basis.”
That cleaning is not a straightforward proposition. Large solar plants can involve hundreds of square kilometres of cells, and manual or vehicle-based cleaning can be risky, labour-intensive, inconsistently effective and involve using large amounts of water.
Ecoppia’s approach, Molad said, is to use autonomous systems that operate using dry cleaning; the company’s robots operate on rails along the solar panels and use a delicate microfibre that only touches the surface of the panel.
“Automation is clearly the only sustainable and cost-effective solution,” he said. “There’s very little dependency on labour and no dependency at all on the cost of the water. And of course, post COVID-19, there’s no risk of lockdowns.”
Challenging tasks
These advantages, however, do not mean the automated approach is without challenges.
“The first one is, of course, the remoteness of the sites,” Molad explained.
“These solutions have to be truly autonomous — and ‘truly autonomous’ means autonomous when there are no people around.
That means that it is crucial to ensure the systems have a high mean time between failures.
“Intervention in their operation is required to be minimal, because we need to minimise the time on-site of personnel due to the high cost,” he said.
“In order to operate the robots from afar and get all the data from the robots and collect and analyse big amounts of data, there needs to be continuous communication between the robots and their operating and control system.”
As a result, the robots need to be highly effective.
“These robots operate day after day after day, for 25 years,” Molad said. “The robot has to remove more than 99.5 per cent of the soiling daily in order to keep generation at its peak.”
And since the landscape around a solar plant can shift over this large timeframe, obstacles can begin appearing, to which the robots must be able to adapt. As a result, cost-effectiveness and compatibility are also important factors.
“We need to be compatible with a lot of different types of structures and configurations that are present in these sites,” Molad said.
“There are many, many different tractor manufacturers coming out every year joining this market, each one coming with a different configuration.”
As well as being adaptable to different kinds of infrastructure, the robots also need to be able to adjust to different environmental conditions.
“They need to withstand high temperatures during the day, low temperatures at night; high humidity that sometimes kicks in after generation hours; and of course, lots and lots of dust. High winds and some monsoons sometimes in Asia in some of our sites,” Molad said, adding that winds and ultraviolet radiation were also concerns.
“To withstand all these weather conditions, we have to select carefully the components that we use for the electronics but also for the mechanics. They have to be long-lasting materials that will [endure] the strong UV radiation.”
It was also important, Molad said, to ensure the robots could operate safely without damaging the solar panels and affecting any warranty protections from the manufacturer.
“We need to make sure that we’re not degrading the panels with the operation of the robot, and also not harming what is called the anti-reflective coating, which is a coating of the panels that is meant to make sure that all radiation comes into the silicon, and is converted into power,” he explained.
Reliability of operation
In solving these challenges, ensuring the reliability of the technology over a matter of decades becomes important. This involves considering not just the cost of the robot over its lifecycle, but also the costs of its operation.
“It’s a product that has to be working and available and have spare parts for 25 years,” Molad said. “We need to use available components in the market that will be there for a long period of time, and also design the robot in a way that will enable upgrades … and use other replacement parts when some parts become [obsolete].”
Molad also highlighted the importance of collaboration, agility in hardware development and drawing on sources such as big data when operating the technology.
“We need to collaborate closely with the market, meaning the other players that are putting their assets in these sorts of fields,” he said.
“We have to collaborate with the site owners. There’s usually an engineering procurement and construction company … planning and building these sites, so you’re collaborating with them closely, understanding the different types of layouts that these sites might have.”
Collaborating with technicians also matters.
“When we sell the robots, we don’t just sell the robots, but you also sell the maintenance service for 25 years,” Molad said. “I urge my engineers to keep on talking to the field guys, getting the feedback and making sure that our next product in development will fit the needs of the field.”
As solar grows in popularity, these challenges will only grow, and Molad has his eye on meeting them.
“There’s currently kind of a global competition in the world for who will build the next biggest utility scale solar PV installation, with records being broken every year,” he said.
“We will keep innovating and creating new solutions to enable this huge scaling of solar sites.”
The article is very timely ; it alerts us to issues that are ignored in the haste to convert to renewables.
Is it possible to provide a similar review on the future potential issues with roof-top solar panels?
I can attest to the effect of soiling. My 17 year old roof top solar panels seemed to have degraded to about 80% of their original capacity. However, they were recently “shotblasted” by a hailstorm followed by very heavy rain and then, on following (currently somewhat rare) sunny days delivered their original rated capacity. This suggests that even after 17 years, the cells themselves have not undergone significant degradation, but soiling is a problem.