Power plant operators continually search for ways to make electricity production and transmission more efficient. With more generation sources coming online, the use of artificial intelligence and machine learning is growing in importance.
The increased digitization of the power generation industry means that artificial intelligence (AI) and machine learning (ML) are becoming synonymous with power generation. AI brings great potential to the future design of energy systems across several applications, including smart grids, electricity trading, efficiency, and the networking of power consumers and generators (Figure 1).
1. There are many uses for artificial intelligence (AI) in the energy industry, and for power generation in particular, including for smart grid applications. AI also can help with operations and maintenance, forecasting, and energy trading. Courtesy: Next Kraftwerke
ML includes a part, but not all, of AI. For power companies, ML provides the opportunity for machines to learn independently, drawing conclusions from their experiences for future operational efficiencies, with the machine solving problems, and reducing operations and maintenance downtime and costs.
Increasing the use of both AI and ML in the power generation sector means making wise use of a growing amount of available data, with a goal of making electricity production and use more efficient and secure. It also means developing ways to more quickly analyze and evaluate ever-larger volumes of data. AI offers many suitable application scenarios to support the integration of renewable energy resources to the power grid, as more utilities adjust their generation fleets to meet climate goals.
“AI is transforming several different industries,” said Colin Parris, senior vice president and chief technology officer for GE Digital, in an interview with POWER, “[including] power generation, information, aviation, and transportation. Today, you have to know how to build a data system and build the analytics.” Parris noted how GE uses the Lean Methodology, which is a way of optimizing the people, resources, effort, and energy of an organization to create value for customers. GE chief executive Larry Culp is a proponent of the system, and GE and other companies have incorporated it into their business practices.
“I grew up in a world where, whenever I did a digital transformation, I had to do a process transformation,” said Parris. “Now, I can go and get all the KPIs [key performance indicators], all the data comes in one spot… for data scientists like me, all the data shows up, and Lean gives me all the business process information.”
Balancing Baseload and Renewable Power
Having volumes of data available, particularly to evaluate the performance of equipment and systems, provides the opportunity to continually improve the production, transmission, and distribution of electricity. It’s become even more important as more renewable energy resources come online.
The balancing act between baseload power, and the intermittent nature of renewables, means utilities and transmission system operators need ways to automate systems to account for fluctuating supply and demand. AI and ML will play an increasingly larger role.
“Promising new sources of renewable energy—solar and wind—are variable in nature. As such, power and wind energy outputs are rarely consistent and uniform, and they depend on a variety of outside factors—intensity of solar radiation, cloud cover, wind speed—that can’t be controlled,” said Dijam Panigrahi, co-founder of Grid Raster, a company that provides computing and network services to a variety of industries, including power generation. GridRaster can create a virtual environment close to real-world settings, and provide realistic product visualization and real-time collaboration, so an effective environment can be created for remote maintenance, repair, and training.
“Machine learning could revolutionize the way we deal with energy,” Panigrahi told POWER. “Its impact ranges across the areas of renewable energy forecasting and smart grids. With ML, for example, we could accurately predict the amount of electricity generated by a wind turbine in the next 36 hours, and therefore, we would be able to effectively transition to renewable energy without destabilizing the power grid.”
“The benefit [of AI and ML] is higher utilization of the existing infrastructure,” said Luke Witmer, general manager for Data Science, Energy Storage, and Optimization at Wärtsilä. “There is a lot of under-utilized infrastructure in the power generation industry. This can be accomplished with greater intelligence on the edges of the network [out at each substation and at each independent generation facility] coupled with greater intelligence at the points of central dispatch.”
Witmer told POWER that AI has a role in the design of virtual power plants, and can be used to model dispatch from other generation facilities. “Today, any power plant [virtual or actual], is designed through a process that involves a lot of modeling, or simulations of what-if scenarios,” he said. “That model must be as accurate as possible, including the controls behavior of not only the new plant in question, but also the rest of the grid and/or markets nearby. As more AI is used in the actual context of this new potential power plant, the model must also contain a reflection of that same AI. No model is perfect, but as more AI gets used in the actual dispatch of power plants, more AI will be needed in the design and creation process for new power plants or aggregations of power generation equipment.”
The real-world applications of AI and ML in power generation are being recognized in a variety of ways, certainly in operations but also as part of workforce efficiencies.
“Globally, experienced power generation operators represent an ever-shrinking workforce whose deep domain know-how and expertise are critical to efficient, safe, and reliable operations,” said Stephen Kwan, director of Product Management for Power Generation/Grid Management at Beyond Limits, an industrial and enterprise-grade AI technology company. Kwan told POWER, “This subject matter expertise needs to be captured, digitized, and made accessible across the workforce in order to ensure the long-term continuity, efficiency, and reliability of operations across the power generation domain. Novel hybrid AI/ML approaches represent a necessary avenue for combining the value of data and domain knowledge to tackle global challenges facing the power generation industry.”
Beyond Limits in December 2020 announced it is working with NVIDIA to advance an initiative for bringing digital transformation to the energy sector. NVIDIA provides AI computing platforms; the company’s graphics processing unit software serves several industries with virtualization solutions.
“AI has the potential to make a major impact on problems facing the heart of the global energy business, but the technology requires high levels of computing power to operate on the level and scale required by many of today’s global producers,” said AJ Abdallat, CEO of Beyond Limits. “With NVIDIA technology support and expertise, Beyond Limits is better positioned to offer faster, more intelligent and efficient AI-based solutions for maximizing energy production and profitability.”
2. A cognitive power plant project in Africa will use artificial intelligence and technology developed by Beyond Limits to monitor the facility’s operations. The technology originated with NASA. Courtesy: Beyond Limits
Beyond Limits in 2019 was awarded a contract by Switzerland-based Xcell, a global financial and minerals development company, to build the world’s first cognitive power plant (Figure 2). The company said the plant “is part of a large-scale infrastructure program to drive core industrial capacity and power economic development in West Africa.” The $25 million deal includes the development of Beyond Limit’s Cognitive Power Generation Advisor, a technology which originated for NASA missions, to monitor operations at a natural gas-fired power plant.
As with anything in the digital realm, cybersecurity concerns are part of the landscape when dealing with increased digitization and the use of data.
“Even with a smart grid, there is a potential source of cyber concern,” said Panigrahi. “A central system that collects data about the energy usage habits of millions of users can emerge as a target for malicious cyberattacks. This could potentially destabilize a grid while also damaging precious consumer data. Power companies must deploy comprehensive cyber defense systems through a regular cadence of perimeter tests in order to have a high level of confidence to protect the system. Additional technologies from augmented reality and virtual reality can aid and assist in these perimeter tests to spot and identify potential threats to the digital perimeter.”
Said Kwan: “Given AI requires a lot of data, it’s vital to incorporate subject matter experts when creating the machine learning/AI models so as to understand the minimum required data set. It’s also important to follow IT best practices to ensure data is isolated from critical infrastructure like the control network, confirm only the least-required privilege is given to the users or applications, and implement logging and audits per industry security measures.”
Kwan continued: “In a typical power generation facility, the control system resides on an isolated and dedicated network. Meanwhile, business users are typically on separate networks that are more ‘open’ for cybersecurity threats. All AI/ML applications require good/clean data for optimal functionality. As such, any data necessary for AI/ML solutions—that reside on the control network—need to be pushed to business networks in a safe and reliable manner. By preventing network traffic into the control network and only allowing data to be pushed from the control network, it decreases the chance of cybersecurity incidents on the control network.”
Kwan noted other aspects of generation related to AI and ML. “AI/ML approaches will add significant value to the ever-growing deployment of large-scale, geographically distributed energy resources,” he said. “The optimal management of trade-offs between meeting demands, ensuring low-risk and reliable operations, and maintaining the integrity of key assets will require AI/ML systems capable of building and supporting holistic models of specific power planning and scheduling, generation, and distribution processes.”
Kwan said the use of AI and ML applications in the power generation and utilities sector will only grow in the coming years. “Integration and adoption of machine learning and other artificial intelligence solutions are already increasing exponentially,” he said. “At this pace, the outlook for advanced technologies points towards AI approaches becoming the gold standard for the future growth of the power generation/utilities industry.”
—Darrell Proctor is associate editor for POWER (@POWERmagazine).