What Will the Power Plant of the Future Look Like?

"Power plants are just robots that don't have a brain yet," Peter Kirk, former chief executive of software company NeuCo, told the BBC recently. As indispensable a resource as they are, these generators have a clear need for smarter technology and greater efficiency, which Kirk's company is aiming to serve with artificial intelligence (AI) solutions.

In the US, the average age of a nuclear power plant is 36 years, and the average age of a coal-powered plant is 39 years. Modernization is key for not only these individual units, but the entire sector as well. The path to digitization will bring more connectivity, more data, more automation, and more efficiency.

While digitization does create challenges around cybersecurity, it also presents opportunities to create power plants that are highly efficient, super smart, more flexible, and better equipped to adapt to new energy trends.

In fact, the US Department of Energy (DOE) has made the smart grid a priority that includes: a "fully automated power delivery network that monitors and controls every consumer and node, ensuring a two-way flow of electricity and information."

AI and Machine Learning

AI combined with cognitive computing will be the brain in the power plant of the future.

Like almost all other industrial sectors, the power generation industry is being disrupted by data-driven AI. Enhanced analytics and AI-enabled algorithms can help identify "anomalous" or "out-of-band" behavior to improve efficiency and help balance energy supply and demand.

Companies like NeuCo, which was acquired by GE Power in 2016, can deploy cognitive computing to consume and learn from data at a rate and level of detail impossible for a human to replicate. Plant leaders can use this information to better manage the minute nuances and changes in a power plant for enhanced performance and efficiency.

Furthermore, as outlined in a McKinsey Global Institute discussion paper titled Artificial Intelligence: The Next Digital Frontier? the emerging technology can discern trends and patterns better than people to more accurately manage energy supply and demand.

Considering the expected exponential growth of electric vehicles in the coming decade, the ability to better balance supply and demand will be paramount. The McKinsey paper says AI can help automate demand-side response and supply prediction, improve electricity production yield, reduce energy waste, improve time of day pricing, provide consumption insights, and much more. According to McKinsey, this can result in a 20 percent energy production increase using machine learning and smart sensors to optimize assets' yield. The firm even believes the digital transformation of utilities could provide such good results that it will mitigate the need to build more capacity in the future.

These computer-based solutions will be teamed with others, including digital twin technology, which is a data-based representation of an asset. These technologies are already enabling asset managers to collect data on every part of their machines to diagnose faults and predict maintenance needs—avoiding guesswork or unnecessary maintenance check-ups.

Energy Storage

Most likely, the energy infrastructure of the future will involve some form of battery storage to accommodate more renewable energy flooding the grid.

In the same way that transmission lines affect whereelectricity is consumed, energy storage will influence when it is consumed, says a report by Deloitte.

Energy storage has the potential to be a major disrupter to incumbent power suppliers, such as coal- and gas-fired power plants. As prices for storage eventually lower, and incentives for renewable energy usage increase, consumers may start to make and consume their own power through the combination of solar and storage. However, energy storage can also work alongside the centralized energy supply and renewable sources to help shave off peak demand and better balance the grid. Deloitte warns that incumbents must prepare for the technology, because "to remain a casual observer is to risk disruption."

For power generators, investment in energy storage could provide an alternative to plant additions. This could enable an AI-incorporated power plant to operate better for longer and more efficiently, avoiding costly start-ups and shutdowns and sidestepping the need to add extra physical capacity as energy demand grows.

Energy storage will no doubt increase decentralized energy. However, a paper by the European Commission suggests that peak increases can be solved where energy storage is available at different levels of the electrical system.

The technology is nascent and not cost effective yet, but there's little doubt that some form of electrical energy storage will play a role in the future energy mix.

Be Future Ready

"Customers' needs are changing," Carlos Nouel, the vice president of National Grid's New Energy Solutions, told Utility Dive. "They [customers] may want to connect their electric vehicle or install solar panels on their roof or maybe they just want to have more data so they can make better decisions around energy," he said.

And so, subsequently, the energy sector is changing. All businesses, including power generators, are essentially becoming digital businesses.

Although digital solutions like AI and machine learning might seem intangible and almost too good to be true, they are usually easily tested, relatively cheaply, and allow users to see results and save money without a huge upfront investment. This is unlike the traditional way operators might look to save money and increase efficiency by investing in hardware upgrades costing millions of dollars. Google recently applied AI technology to reduce its total data center power consumption, which translated to millions of dollars in savings.

Partnering with start-ups and digital-based companies like NeuCo can help operators learn how to save and also test the waters, so to speak.

Therefore, the power plant of the future will be fully connected, more efficient, and operational more hours of the year. It will process more data, be more flexible, and still play a vital role in the future of the global energy mix.