Access to clean, cheap, and reliable energy is at the very heart of development. Therefore, a lack of energy access is a fundamental impediment to progress that impacts health, education, food, and livelihoods. Universal access to affordable, reliable, and sustainable energy is the primary goal in emerging markets and advanced economies.
Yet, it will remain a mere goal unless innovative solutions and modern technologies overcome numerous energy-related obstacles, such as diversification and decentralization of energy production, changing supply and demand patterns, lack of sufficient power generation, lack of analytics needed for optimal management, poor transmission and distribution, affordability, and climate concerns.
Artificial intelligence (AI) and related technologies have the potential to address these challenges, which are more acute in emerging market nations, where efficiency issues are particularly problematic. These technologies can improve power management, efficiency, and transparency, and cut energy waste, lower costs, and accelerate the use of clean, renewable energy.
If designed carefully, AI systems can be handy in the automation of routine and structured tasks in the power grid, improving the planning, operation, and control of power systems. Thus, in many ways, AI technologies are closely tied to providing clean and cheap energy, which is essential to development.
Thanks to AI – the power sector has a promising future with the advent of AI-managed smart grids that allow two-way communication between utilities and consumers.
Smart grids, embedded with an information layer of advanced sensors and smart meters for data collection, storage, and analysis, enable a real-time and seamless interaction between multiple remote points and components across the grid, so they can better respond to quick changes in energy demand or urgent situations. It creates a current, precise, and integrated view of the entire power system, facilitating better grid management.
Paired with powerful big data analytics, cloud computing, and the internet of things (IoT), these smart-grid elements can improve the reliability, security, and efficiency of electricity transmission and distribution networks. This data analysis can be further used for various purposes, including fault detection, predictive maintenance, power quality monitoring, and renewable energy forecasting.
Fault prediction
Fault prediction is one of the major AI applications in the energy industry, where equipment failure is common, with potentially significant consequences. AI, coupled with sensors, can monitor equipment and detect failures before they happen, thus saving resources, money, time, and lives.
For example, predictive diagnostics are currently used to predict problems that could potentially shut down geothermal power plants, yielding steady energy output. IoT and AI can optimize preventive measures such as chemical agent sprays to avoid turbine shutdowns. It improves the efficiency and reliability of geothermal power plants.
Maintenance
National Grid in the UK has recently turned to drones to monitor wires and pylons transmitting electricity from power stations to thousands of homes and businesses. Equipped with high-resolution cameras, these drones are particularly useful in fault detection due to their ability to cover vast and difficult terrain. They cover 7,200 miles of overhead lines across England and Wales. AI is then used to monitor power assets’ conditions and determine when they need to be replaced or repaired.
Energy efficiency decision making
The digital transformation of home energy management and consumer appliances, using smart devices such as Amazon Alexa, Google Home, and Google Nest, enables customers to interact with their thermostats and other control systems to monitor their energy consumption. They allow automatic meters to use AI and optimize energy consumption and storage. For instance, it can trigger appliances to be turned off when power is expensive or electricity to be stored when power is cheap, or solar rooftop energy is abundant.
Furthermore, in deregulated markets like the United States, where consumers can opt for their energy providers, AI empowers consumers by allowing them to determine their provider based on their energy source preferences, household budget, or consumption patterns.
Notably, researchers at Carnegie Mellon University developed a machine learning system named “Lumator” that combines the customer’s preferences and consumption data with different tariff plans, limited-time promotional rates, and other product offers. It then provides recommendations for the most suitable electricity supply deal. Since it becomes familiar with the customer’s habits, it can automatically switch energy plans when better deals are available without interrupting supply.
Disaster recovery
When Hurricane Irma struck South Florida in 2017, it took ten days to restore power and light, instead of 18 days, which was needed for the region to recover from the previous hurricane, Wilma. This time reduction was due to technologies such as AI predicting power availability and ensuring it is delivered where it is most needed without negatively impacting the system.
Furthermore, AI systems can improve assessments of damages and optimization of decision making thanks to a faster access to imagery and information—within the first 12 to 24 hours—after the disaster has subsided.
Prevention of losses due to informal connections
Losses due to informal connections constitute another challenge for the power sector. AI can be used to spot discrepancies in usage patterns, payment history, and other consumer data to detect these informal connections. When combined with automated meters, it can also improve monitoring, optimizing the costly, and time-consuming physical inspections.
For instance, Brazil has been suffering from a high rate of nontechnical losses that include informal connections and billing errors. The University of Luxembourg has developed an algorithm that analyzes information from electricity meters to detect abnormal usage. The algorithm managed to reveal problem cases at a higher rate than most other tools when applied to information over five years from 3.6 million Brazilian households.
Conclusion
Though AI holds considerable potential to improve power generation, transmission, distribution, and consumption, there is a need to educate the AI industry more deeply on the power sector aspects. Besides, there are regulatory restrictions in cloud-based applications, which are widespread and central to AI solutions. Integrating different data sources and ensuring representativeness given the diversity within the data is also challenging. Like other sectors that are increasingly applying AI technology, if the power sector can address all these challenges, AI plays an important role in the energy sector.
