People are exploring how to use artificial intelligence (AI) in their work. Such efforts have shaped designers’ attempts to apply AI in energy industry projects, including those using hydropower. These are emerging opportunities in the early stages, but they are exciting.
Working With Adaptable Digital Twins
What if designers could be more proactive in suggesting solutions that will keep hydropower systems running smoothly for longer? That is the goal behind a joint project from two United States centers.
Researchers from Oak Ridge National and Pacific Northwest National Laboratories are collaborating to build a digital twin framework for hydropower optimization. The framework would allow users to create and run experiments before implementing their ideas in practice.
One of the most compelling characteristics of some AI algorithms is that they adapt to changing conditions and show improved performance through ongoing use. Details about the digital twin framework suggest it would have artificial intelligence working in the background.
In addition to collecting operational data, the tool would identify probable issues and provide suggested fixes. Additionally, this project’s coverage indicates the digital twin would have autonomous capabilities that allow it to learn from and reason about the surrounding environment.
The digital twin supports hydropower improvements by providing targeted resources for designers, operators and maintenance professionals, assisting these parties in making the most appropriate decisions while influencing projects. Moreover, testing different possibilities in digital twins lowers costs by reducing expensive failures or oversights.
Deploying Hydropower While Minimizing Environmental Impacts
An ongoing challenge concerns how people can develop efficient hydropower projects without causing excessive environmental damage. Succeeding could help the public view hydropower as critical to future sustainability. More than 60% of today’s renewable energy comes from it, showing good progress.
Cornell University researchers believed AI could highlight how to minimize hydropower’s environmental damage. They started with a daunting project involving the Marañón River and its numerous dams slated for construction in Peru and Ecuador. That body of water stretches for nearly 1,100 miles and is one of the Amazon River’s primary headwaters.
However, as the team’s work progressed, they became interested in applying AI to the entire Amazon basin, which comprises more than a third of South America. That effort and the corresponding academic paper involved dozens of co-authors from three continents who were associated with more than 24 educational institutions.
The collaborators examined hydropower optimization in the context of more than 350 dams proposed for the region. Their AI application analyzed six factors, including emissions, sediment transport and fish diversity. The algorithms also examined characteristics of 158 existing hydropower projects, finding missed opportunities so designers, engineers and other professionals could avoid past mistakes and adverse effects caused by insufficient planning coordination.
Studying the results allows professionals to identify and eliminate the lowest-quality dam configurations and sites that would harm the environment the most. Artificial intelligence also gives basin-wide analyses, which is particularly important since what is best for a single area may not be optimal for the entire river. The focus on specific elements before providing broader views supports project developers in recognizing their decisions’ potential effects and reacting accordingly.
Performing Maintenance Inspections
Many practical applications of AI in energy industry settings center on risk management. Maintenance technicians look for early signs of trouble, flagging those concerns so decision-makers can remove or reduce threats before they cause faults or outages.
These workers also suggest best practices to counteract factors that could compromise a hydropower system’s performance or reliability. For example, corrosion, strain and moisture can damage power transmission cables. Flexible conduits are among the best safeguards for outdoor applications such as hydropower systems. However, the size, scope and number of finished or in-progress projects make it difficult for technicians to manually inspect all applicable sites to find problems. Artificial intelligence fills the labor gap.
Japan’s J-POWER operates all the country’s transmission lines and has made significant investments in AI applications for hydropower. The enterprise’s leaders designated Shimogo Power Station and the surrounding area as a digital-focused zone dedicated to hydropower maintenance improvements.
In one example, the business developed a proprietary AI tool that collects data from hydropower assets, examines it for abnormalities and alerts the appropriate staff. Additionally, executives tapped a firm to assist with developing and implementing robots that patrol critical areas around hydropower infrastructure. Previously, the energy provider was the first in Japan to use a four-legged inspection robot at its hydroelectric power plants. The machine could keep operating during disasters, becoming an essential part of the organization’s preparedness.
Some hydropower systems also have complementing equipment to check, such as floating solar panels. This equipment can store energy for later, too, keeping dams operational during low water-level periods. A 2024 study showed such components could meet entire countries’ electricity needs, making them worth further exploration. Scientists reached that conclusion after reviewing data from 68,000 of the world’s lakes and reservoirs.
Upending Conventional Labor Practices
Using AI in energy industry applications requires capitalizing on human expertise and ingenuity while understanding how the technology excels. Artificial intelligence handles many repetitive tasks well, but it cannot equal what people achieve during many creative or problem-solving endeavors.
One ambitious endeavor comes from China, where AI-powered autonomous construction equipment, robots and 3D printing will build a nearly 600-foot-tall power plant with no human workers contributing to the daily labor. However, people are still heavily involved because scientists give ongoing input to test and create this unusual construction process. Researchers hope the high-tech collaboration will free individuals from dangerous and intensive work, giving them more rewarding opportunities.
This project’s lead researcher and his team conceptualized this construction approach about a decade ago, envisioning entire construction sites becoming 3D printers that work without human oversight. The process begins by splitting a computerized model of the dam into slices. Next, robots work in teams to create each layer. Automated trucks and excavators are also in the vicinity, identifying materials and moving loads of them into the right areas.
AI calculates the optimal vehicle path, and the goods move into that area. Later, bulldozers and pavers arrive to create new layers. The artificial intelligence can also sense ground vibrations, determining overall traffic in specific parts of the site.
Although this is an unusual hydropower optimization example and not one someone could quickly replicate elsewhere, it is nonetheless valuable for showing the possibilities and inspiring leaders to automate more construction steps when possible.
These fascinating cases show open-minded professionals can apply artificial intelligence to get impressive results. Such examples will remain valuable, especially as more companies plan and develop hydropower projects.
