Since the first iterations of robots and machine learning appeared, their possibilities have captured our imaginations and our darkest fears. The menacing potential of robots and machines to replace humans has always seemed like a science fiction scenario that wasn't entirely fiction.
Ironically, this technology can do just the opposite, by helping humanity overcome a very real and existential threat in the form of climate change. In fact, artificial intelligence (AI) and robotics make a great team as society attempts to transition to clean and renewable sources of energy generation.
Transforming the nation's energy infrastructure requires numerous changes, not just to the fuel sources but to the myriad of support systems and equipment required to achieve this transformation. Advanced digital and mechanical innovation can facilitate these ancillary needs in several ways.
Power Production
The first step in renewable energy generation is development. AI and robotics are helping renewable developers in many ways. Sites must be identified and evaluated, and the necessary equipment must be manufactured and installed before any power can be generated.
Just as robotics are doing in other industries, they are improving the manufacturing process for solar panels, wind turbines, and other renewable generating equipment. The role of robotics in manufacturing has been expanding, as robots perform tasks more quickly, efficiently, and safely than humans. More companies are also turning to robotics when they need to address a shortage of qualified labor in manufacturing.
The California-based manufacturer, Orbital Composites, specializes in “reducing blade costs and footprints by cutting transportation expenses.” Using a grant from the US Department of Energy (DOE), the company is teaming up with Oak Ridge National Laboratory and the University of Maine to develop a process using 3D printing and robotics to manufacture wind turbine blades on the site of their installation. The process will streamline the manufacturing of turbines and eliminate the logistical challenges of delivering equipment to remote, inaccessible locations.
After power generating equipment has been manufactured, developers use data analytics, algorithms, and modeling—all features of AI—for more effectively evaluating sites for the development of renewable sources, like wind power, solar, and geothermal.
DroneDeploy is a software company that employs so-called “digital twins” or computer renderings to help companies evaluate sites. Using drones, cameras, robots, and software, the company captures data and produces maps and modules. Solar developers use this information to make informed assessments of their sites and implement a layout of panels that maximizes electricity generation.
Geothermal power also faces challenges in site evaluation that can be addressed with AI and robotics. The selection process is typically very labor intensive. Developers analyze geological surveys that are conducted by humans relying on traditional methods.
AI can dramatically improve the speed, reliability, and accuracy of this process by analyzing vast amounts of geological data and identifying suitable reservoirs of geothermal resources.
AI can dramatically improve the speed, reliability, and accuracy of this process by analyzing vast amounts of geological data and identifying suitable reservoirs of geothermal resources. This saves companies money and time, enabling them to effectively select sites with the greatest potential for development.
Inspection and Maintenance
Once renewable sites are developed and utilized, they must be inspected and maintained. AI and robotics can help utilities improve the safety and efficiency of these processes.
Wind turbines and solar farm panels are typically installed in locations that are far from densely populated areas and often difficult to access. Even when located in more accessible places, wind turbines still require teams of humans to scale dangerous heights using ropes and other suspended equipment to make visual inspections. Likewise, geothermal and wave generation resources pose accessibility challenges to humans due to the uniqueness of their locations.
Utilities can now reduce time, costs, and safety risks by using smart technology to perform inspections and maintenance.
Utilities can now reduce time, costs, and safety risks by using smart technology to perform inspections and maintenance.
Tim Lichti is the co-founder and CEO of Swap Robotics, an Ontario, Canada-based company that builds robots designed to provide yard maintenance on solar farms. The company’s “solar vegetation robots” provide a much-needed service around the panels. The mobile units drive through the rows of panel arrays, cutting grass and weeds before they interfere with solar generation.
According to Lichti, “the outdoor world needs to be maintained.” He explains that “after the solar farm is built, the biggest expense that is ongoing is cutting the grass and the vegetation.”
Drones can also fly into remote locations and inspect turbines or geothermal wells. They can scan, take high-resolution images, and capture data in a fraction of the time it takes a team of human inspectors to perform the same tasks. The information they gather helps detect issues that might go unnoticed by the human eye. Drones can also inspect turbines, panels, and other equipment for damage after a storm, wildfire, or other natural disaster when human access is impeded.
Robotic technology aids in the inspection of large solar farms. Mobile robots, which resemble a toy car that a toddler would love to drive, can travel an entire solar farm, inspect materials, take images, and capture data on panels and arrays more quickly and efficiently than humans. Some robots are equipped with highly sophisticated thermal cameras to detect abnormal temperature fluctuations. These signal issues or defects in the panels that must be addressed.
Robots are also deployed to provide security for solar farms. They move around the perimeter of the farm and send live images back to monitors in a control center where humans can respond to security issues. Because they are mobile, these robots cover much more ground and many more angles than stationary monitors and can help utilities avoid theft of expensive equipment.
Would-be thieves aren't the only interlopers utilities need to keep away from their renewable generating facilities. Birds can be a major nuisance and safety hazard. In many cases, the birds themselves are a threatened or endangered species, presenting utilities with the added challenge of having to protect their equipment without harming or killing the culprits.
Here, too, robotic technology and AI are helping utilities come up with a solution. The Edge Company is an Italian-based firm that uses artificial intelligence to help wind farm operators detect the presence of flocks of birds to avoid collisions and disruptions in their operations. The S9 Bird Control Robot by the California-based SMP Robotics scares away flocks from solar farm panels to help keep the panels clean and power generation uninterrupted.
Robotics can also assist in extreme climates, where freezing temperatures can disrupt and damage wind turbines. The Latvian company, Areones, uses a specially designed drone to lift heavy materials to efficiently clean, spray, and de-ice frozen turbine blades.
Artificial intelligence and machine learning help with maintenance by using data analytics and algorithms to identify, predict, and isolate problems.
Finally, artificial intelligence and machine learning help with maintenance by using data analytics and algorithms to identify, predict, and isolate problems; schedule appropriate maintenance; and accurately time generation. For example, they can analyze weather and tide information to help schedule wind or tidal power generation, maximize output, and minimize unnecessary strains on the system.
Grid Management
One of the biggest challenges for the growth of renewables is achieving and maintaining a balanced integration with the existing distribution grid.
Some renewable sources, such as wind and solar, are highly intermittent. They can generate power only when the resource is plentiful, e.g., when the wind blows and the sun shines. These times don't necessarily correspond to the times of peak demand.
Furthermore, renewable power is often generated by small facilities that are located closer to the consumer rather than on or near large sites, like a coal-fired power plant. These distributed generating facilities pose unique challenges to grid operators.
Traditional power generation from coal-fired plants, for example, take time to achieve peak generation or “ramp up.” Renewables can be used to compensate for this delay, but only if they are plentiful and in ready supply.
These and other unique characteristics factor into the management of power generation, demand, and distribution on the grid. Operators must take steps to maintain a balanced or level flow of energy, ensuring adequate supply, avoiding disruptions, and preventing spikes.
Grid operators must take steps to maintain a balanced or level flow of energy, ensuring adequate supply, avoiding disruptions, and preventing spikes.
AI and robotics are helping the industry achieve this goal. The solution revolves mostly around the gathering, analysis, and application of data. This synthesis of digital information has also created its own growing vernacular of new terminology.
For example, the combined effect of data processing is now commonly referred to as the “smart grid.” A smart grid processes information about demand, generation, and distribution much more quickly and efficiently than humans, making the grid run more smoothly and reliably, without disruptions.
One feature of the smart grid is predictive analytics. Through computer analysis of information about renewable resources and weather patterns, utilities can make proactive decisions about when power is most likely to be generated, how much will be generated, which assets will generate it, and how the production will intersect with expected demand. Through this analysis, grid operators can maximize the use of renewable resources to ensure smooth operation on the grid.
One feature of a smart grid created by AI is the so-called “self-healing grid.” Centralized computer systems gather information from remote sensors and equipment that communicate with one another over the internet, otherwise known as the “Internet of Things” (IoT). This processing of information from multiple remote sites enables utilities to make continuous “self-assessments” to quickly detect issues and avoid service disruptions.
Traditionally, utilities have had to rely on customer complaints to identify issues, then send out teams to make repairs. Instead, the self-healing smart grid allows them to detect, and sometimes even predict, problems in real-time and reroute power to avoid outages and the need to deploy teams of service repair personnel.
The smart grid has also created a new, hybrid type of consumer: “Prosumers” are utility customers who consume and produce power.
The smart grid has also created a new, hybrid type of consumer. “Prosumers” are utility customers who consume and produce power. Owners of solar panels and electric vehicles (EVs) are becoming prosumers.
The owner of a home with rooftop solar panels consumes electricity drawn from the grid when it is needed. At other times, the owner may send excess power back to the utility over the grid. This will happen during the day when the sun is at its peak and the panels are busy generating electricity, but the home is not drawing much power because the owner is away.
EV owners may also be prosumers. At times, they will draw power from the grid to charge their vehicle. At other times, the utility may draw power from the charged vehicle and send it to other sites where it is needed. In this way, the utility is using the vehicle like a storage device that supports the grid with backup power.
Smart grid technology makes prosumers possible and gives utilities and grid operators the ability to manage them effectively. This allows them to manage power from renewables more effectively to ensure smooth distribution over the grid.
AI and robotics are also empowering so-called micro-grids. These are small subsets of grids that service customers in a limited and defined area, often ones that are remote and inaccessible. Microgrids provide their own generation, typically through one or more sources of renewable power. They support the smooth operation of larger grids and often provide security and protection against various forms of natural disasters that can disrupt service in the areas. In the same way that grid operators are using the technology, micro-grids are also using AI to manage generation and distribution of their limited resources. Micro-grids are a useful tool for grid operators, and AI is making them more feasible in various locations. (For more on solar micro-grids see also “Power for the People”—How Solar Mini-Grids Help the Disadvantaged, April 2022, theearthandi.org.)
Conclusion
Stephen Hawking has been quoted as saying that “AI is either the best thing that ever happened to humanity, or the worst.”
Scary images of metallic evildoers bent on the destruction of their creators make for good movies and comic books. The reality is much more nuanced and constructive.
The development and use of AI and robotic technology does pose questions of an existential nature. Far from being a harbinger of the end of the human race, technology can and is being harnessed to help save it.
Climate change is perhaps the greatest threat ever posed to humanity’s continued existence on the planet. AI and robotics are being used in many ways to help humanity face this threat.
*Rick Laezman is a freelance writer in Los Angeles, California, US. He has a passion for energy efficiency and innovation. He has covered renewable power and other related subjects for over ten years.