Prior to the arrival of large language models (LLMs) as seen with technologies like ChatGPT, the use of AI within the world of energy, particularly energy utilities and the electrical grid was defined by the massive data sets derived from the connectivity of the grid; the acceleration of renewables and grid technology that required the use of AI for management of variable power generation; and other similar trends.
Google ‘big data and energy‘ and you can get a sense of what was happening. The phrase ‘smart grid’ was with us for years – and much of the ‘smart’ within the grid already involved AI.
Today? There is a lot more yet to come.
Understanding the Smart Grid and the Future of Energy
I described the future of energy in this way within my Big Future series:
In essence, we are headed to a world of ‘connected energy,’ which was what I covered within that series – and there’s where most of the impact of AI is to be found. Understand the ‘connected energy’ trend, and you’ll understand the real role of AI in the energy/utility industry.
Trend: The Arrival of the Era of Connected Energy
Bottom line: energy in our electrical system used to be one-way and was mostly from carbon. We are now on the edge of an era in which it is becoming a two-way system, with generation primarily based on renewables, distributed, connected, intelligent, and part of a super big hyperconnected microgrid with a lot of batteries. And AI plays a huge role already and an even bigger role in this new energy grid going forward.
For lack of a better term, I called it intelligent, connected smart energy, and it takes us into a world in which our relationship with energy changes – and leads to the birth of new billion-dollar industry opportunities!
Essentially, we are building a new energy grid, launching new industries, changing entire mobility platforms (automotive and trucking), and setting a path to a cleaner energy future. Much of this is currently happening in isolation, but over the decade to come, a lot of connectivity of these disparate platforms and trends is going to occur. It’s already underway as smart innovators build out a fascinating new future.
In a nutshell:
- carbon is over as solar, wind and renewable energy comes to dominate our future
- commercial, industrial, and residential properties are going to see the arrival of more battery storage technology to store this energy, smoothing out the dips and spurts that come with this generation
- most new cars and trucks will be electric within a decade as the era of gasoline and diesel energy comes to an end
- the batteries in those cars and trucks will come to be a part of the energy storage solution in these locations
- on top of that, most homes are getting smart appliances with the potential for connectivity and better management of their energy usage
- at the same time, factories and industrial facilities are getting a lot of interconnected devices through iiOT (Industrial Internet of Things) technology that also allows for intelligent energy management
- the home thermostat, which has gained intelligence through devices like the Nest and ecobee thermostats, is about to take on a larger role in managing this internal home energy grid
- at the same time, sophisticated platforms for energy management in industrial, commercial, and factory locations will do the same thing
- all of this provides a little energy grid within the home and in these industries/commercial facilities, providing a form of energy independence
- and a key component is that any excess energy within this grid will be fed back out into the large utility grid to provide it with additional power, transitioning it from a one-way to a two-way systems
- this grid connectivity also provides the opportunity for local community microgrids – you’ve got a grid, I’ve got a grid – let’s connect our grids
I told this story when booked by the CEO of the California energy giant PG&E a number of years ago:
What it represents
Within this trend are several massive opportunities. I’ve often liked to point out that in many industries, this is the big disruption:
Companies that do not yet exist will build products not yet conceived, with materials not yet in existence, using methodologies not yet developed – and these will be sold to consumers who do not yet know that these new products and services will become a critical part of their life.
It’s in that type of thinking that this new industry is unfolding. There are so many different opportunities it is staggering:
- grid-connected energy storage. Essentially, great big batteries store energy generated by the distributed energy resources in the system. Solar is only generated while the sun is up, and wind energy only happens on a windy day – but grid-connected energy storage solves this problem. There is a huge amount of investment, innovation, initiative, and idea generation going into this – as I explained many years ago, “The Future of Just About Everything is all About Batteries.”
- HEMS or home energy management systems – We will see the evolution of a “home energy ecosystem,” in which later versions of Nest and ecobee will not only let individuals adjust and manage the temperature of their homes but will manage and monitor the storage of electricity generated on their rooftop solar panels, as well as mediate and manage the sale of that energy back to the local power company or a local power grid. In addition, HEMS devices will help to automatically deliver cost and carbon savings from intelligent appliances and other in-home electrical devices, offering up power-saving recommendations.
- new energy management software platforms: there’s a lot of software to make this super-intelligent microgrid system work. With that in mind, consider ampOS, an energy management platform that helps to assess the health of industrial-scale battery packs, vehicle-to-grid connecting, and leasing groups for warranties and insurance
- “EaaS” – or “Energy as a Service.” This new grid involves not only new energy sources but new energy uses including electric vehicles. In that context, consider UgoWork, a Quebec, Canada company that has an AI-based software platform that provides for easier management of the large-scale fleet deployment of batteries
- battery recycling: most electric vehicle batteries are designed for a lifespan of about 12 years before they lose efficiency, but they can be repurposed and reused in microgrid energy storage solutions for a ‘second life’. One company, Connected Energy in the UK, has equity investment from Volvo, Caterpillar, and others, who recognize the emerging secondary market for used batteries. Connected Energy will specialize in taking old car batteries and building them into large-scale commercial energy storage systems.
These are just a few of the components of what is underway. All of this involves the development of a lot of communication protocols, technical standards, interoperability methodologies, and security standards. Suffice it to say, we will wake up within a decade and realize a vast number of new industries and companies were born.
Where are we now? There’s a LOT of innovation and research happening at a furious pace.
- we’ve got a lot of solar energy that has come on stream, but the process of storing that solar energy in a vehicle takes a lot of research. Obviously, organizations like Tesla are leading the way, but other car companies will get involved as the ‘race to electric’ continues unabated. Expect a ridiculous amount of innovation around battery storage technology.
- there is a lot of small-scale or local (i.e. airport or industrial facility) microgrid experimentation and implementation underway, but it is a different ballgame when hundreds of millions of sources and facilities are added. This adds scale to the trend.
- there are big bold initiatives happening. The massive Saudi futurist city initiative, NEOM, has bold goals including intelligent water, wastewater, and energy systems all of which are related to this trend. Big ideas beget big opportunities.
- standards are fast emerging, with the Matter initiative that was just finalized at the Consumer Electronics Show (CES) in Las Vegas. Samsung, GE, Amazon, and LG are among companies with devices and appliances supporting Matter, which allows interconnectivity across devices and management of those devices by HEMS technology
This is just the tip of the iceberg!
The current reality?
The scale of this architectural change is massive.
- Right now, there are only 7 major North American grids. There are about to be hundreds of millions of new inputs to the grid. The same is happening in Europe, Asia, and elsewhere. That’s a lot of innovation and investment that will occur right there.
- There are few electric cars – there are about to be a whole bunch more.
- There’s a fair bit of wind and solar about – but there is a lot more yet to come, particularly as the cost of solar continues to plummet.
- There are few home batteries about – but there will be many more. There aren’t many batteries in the grid – but the scale is coming quickly.
- There is a little bit of intelligent energy technology – but the innovation around this is happening at a furious pace!
- The larger story about connected energy – and why this is such a big trend – is that it is part of a larger story, and that involves the future of infrastructure. I recently caught that on stage in a keynote for the leadership team of one of the world’s largest construction companies.
It’s a trend, it’s happening, and it’s huge!
Key opportunities for AI in the energy/utility sector
Where does this lead us? There are several trends that are happening all at once and which are interconnected, to pardon a utility industry pun.
- global electrical demand is expected to double by about 2050
- the new generation will involve a lot of intermittent power sources such as wind, solar, and other renewables
- it’s difficult to predict production from these systems due to t intermittency (i.e. wind patterns)
- much of the energy grid is transitioning from one-way to two-way with the input of millions of small renewable power sources
- demand on the grid will increase with other trends, i.e. electric vehicles and always-on technology, i. e. iOT
- the renewable two-way grid is also being built with a lot of temporary storage with the introduction of batteries
- all of this will involve an extensive balancing of this new, two-way distributed grid with AI.
In effect, AI can help in managing demand and supply between the network and the source, and AI algorithms will revolutionize the way we generate, distribute, and consume energy from renewable sources such as solar, wind, and hydro because we. can use machine learning to make wind power sufficiently more predictable, or use AI to better determine the best locations for solar power generation.
Power companies increasingly need AI and software to predict when and where power will be available—and automation to make sure power gets to where it’s needed, when it’s needed, without overloading the system.
The Key to Keeping the Lights On: Artificial Intelligence; Power companies are turning to AI, drones and sensors to curtail outages, save money and help operate an increasingly complex electricity grid.
The Wall Street Journal Online, 7 February 2020
The transition of other trends – ie. electric cars – accelerated the demand for the role of AI in this new smart grid.
When a larger percentage of people own electric cars, for example, evening charge cycles when people get home from work will cause a shift in usage that grids will have to adapt to. Keeping grid equipment from breaking down becomes an even more complex task when solar is feeding power back into the system.
“You’ve got to generate enough power to support ongoing demand, but then as renewables come on and offline it gets more complex, and it becomes much more critical to have software that can manage all these renewables,” says Pat Byrne, chief executive of General Electric’s digital business. The company sells analytics software to help manage such shifts, as well as something called “digital twins”—or simulations that help predict how equipment would behave in different conditions. With such tools, power system operators can dynamically control power-generating wind and gas-fired turbines, making sure no individual turbine runs too hot when electricity demand goes up, thus preventing wear and tear.
Batteries, which can store energy from wind and solar, are another area where AI could play a role in the power grids of the future. One company, Stem Inc. of Millbrae, Calif., helps corporations in the U.S. source battery systems, and then uses software called Athena that learns companies’ consumption patterns and autonomously decides when to use battery power to avoid consumption when prices are the highest—in the middle of a hot day, for example.
As markets for electricity become more open, such software could make money for battery owners through a form of arbitrage, charging up when electricity is cheap and disbursing it to the grid when it’s more expensive.
The Key to Keeping the Lights On: Artificial Intelligence; Power companies are turning to AI, drones and sensors to curtail outages, save money and help operate an increasingly complex electricity grid.
The Wall Street Journal Online, 7 February 2020
AI was absolutely necessary with the arrival of smart grids:
The level of grid autonomy, reliability, and optimality of operation depends on the effectiveness of its monitoring, control, and protection methods. Today, smart grids are characterized by a bi-directional flow of electricity and information and leverage distributed computing and communications to deliver real-time information and enable near-instantaneous corrective actions.
Smart grids are typically characterized by a high penetration of distributed generation, consumption, and metering devices. These include wind, solar, hydro, tidal, and thermal generation units, electric vehicles, microgrids, residential smart meters, etc. Control and optimization of all these elements require exchanging an enormous volume of data and metering information that should be ingested, inferred, and reflected upon. In this context, big data analytics, in particular artificial intelligence methods are a key enabler of fully autonomous smart grids.
The automation being applied to smart grids is similar in concept to the network management and operations support systems that were applied to the telecommunications network in the 1970s and 1980s. However, applying IT and communications technology to the grid is not straightforward because it must account for constraints that did not exist in automating the telecommunications network. It is desirable that the unique challenges of the grid are addressed in the automation of the grid.”
BluWave Inc. Patent Application Titled “Systems And Methods For Distributed Hierarchical Artificial Intelligence In Smart Grids”
9 February 2021, Information Technology Newsweekly
Going Forward – The Accelerating Role of AI
The thing is – we’ve barely scratched the surface of the use and implementation of AI in the energy sector; given its role so far, the sudden acceleration of all forms of AI will mean that its role will only become more profound. My online research service led me to a small sampling of articles on the myriad of ways in which it might be used.
Wind farm weather prediction
A study published in the Journal of Applied Ecology tested an IdentiFlight International LLC optical system that tells wind farms when to idle specific turbines by detecting incoming birds whose species it can identify through machine learning
Portland General Electric is slated to pilot a leading-edge artificial intelligence platform, tech company Utilladata announced March 23.
The smart grid chip technology is a new distributed AI platform installed alongside electric meters, intended to integrate distributed energy resources such as solar, battery storage, and electric vehicles — enhancing the resiliency of the grid.
The Rhode Island-based Utilidata’s president and CEO is Jess Melanson.
“Modernizing grid infrastructure in time to meet essential decarbonization goals requires transformational leadership,” Melanson said. “Utilidata is proud to work with Portland General Electric, an innovative partner who understands the value of making investments to not only solve today’s challenges, but to prepare for the dramatic changes the grid will face in the coming years. We are inspired by PGE’s commitment to innovation and look forward to working together to transform grid operations and better serve PGE customers.”
The smart grid chip is powered by the NVIDIA Jetson platform of accelerated computing to collect and analyze large amounts of real-time granular data at the edge of the grid. Altogether, it’s expected to seamlessly add more clean energy, reduce power outages, enable quicker storm recovery and lower the cost of grid operations — while allowing PGE and partners to develop custom solutions.
PGE to pilot AI smart grid chip
5 April 2023, Portland Tribune
Hybrid energy systems
What are hybrid solar PV-wind systems?
A hybrid wind-PV power system is a renewable energy system that combines wind turbines and photovoltaic (PV) panels to generate electricity. The system can be designed to operate in grid-tied or off-grid modes, depending on the application.
Controlling PV-Wind system using AI and ML
Advanced control algorithms based on AI and ML can help achieve effective control of a hybrid wind-PV power system. These algorithms can optimize the power output of the system by predicting the energy generation from both sources and adjusting the system parameters to ensure a stable and consistent power output.
One way AI and ML algorithms can be used is through predictive analytics. These algorithms can analyze large amounts of data from renewable energy sources, weather patterns, and energy usage patterns to predict future energy demand and supply. This information can then be used to optimize the operation of the hybrid wind-PV power system to match energy supply with demand, reducing waste and maximizing energy efficiency.
Another way AI and ML algorithms can be used is through control strategies that utilize data-driven models of the wind-PV system. For example, an AI-based control algorithm can be designed to predict the power output of a wind turbine based on weather conditions such as wind speed and direction. Similarly, a ML-based control algorithm can be designed to predict the power output of a PV panel based on factors such as solar irradiance and temperature.
These predictions can then be used to optimize the operation of the wind-PV system. For instance, if the predicted power output of the wind turbine is lower than expected due to low wind speeds, the control algorithm can adjust the system to compensate for the shortfall by increasing the output from the PV panels. Similarly, if the predicted power output of the PV panels is lower than expected due to cloud cover, the control algorithm can increase the output from the wind turbines.
Artificial Intelligence Can Improve Renewable Energy Production, Report Shows
12 April 2023, Morocco World News
Better monitoring and management of wind farms:
Here we conducted a feasibility study with a focus on monitoring these effects by utilizing different machine-learning methods. A multi-source dataset for a study site in the North Sea is created by combining satellite data, local in situ data, and a hydrodynamic model. The machine learning algorithm DTWkNN, which is based on dynamic time warping and k-nearest neighbor, is used for multivariate time series data imputation. Subsequently, unsupervised anomaly detection is performed to identify possible inferences in the dynamic and interdepending marine environment around the offshore wind farm. The anomaly results are analyzed in terms of location, density and temporal variability, granting access to information and building a basis for explanation. Temporal detection of anomalies with COPOD is found to be a suitable method.”
According to the news editors, the research concluded: “Actionable insights are the direction and magnitude of potential effects of the wind farm on the marine environment, depending on the wind direction. This study works towards a digital twin of offshore wind farms and provides a set of methods based on machine learning to monitor and evaluate offshore wind farm effects, supporting stakeholders with information for decision making on future maritime energy infrastructures.”
“However, predicting wind power is not easy due to the nonlinearity in wind speed that eventually depends on weather conditions. To reduce these issues improved forecasting models have been used to get the correct results and improve the performance and stability of the power system and thereby its reliability and security. In this work, two models are used to predict the ‘Output of Wind Turbine’ to improve the prediction accuracy of short-term wind power generation. The two models namely the Gated Recurrent Unit (GRU) from the deep learning model and Autoregressive Integrated Moving Average (ARIMA) from Statistical Learning.
Energy – Wind Farms; Recent Findings in Wind Farms Described by a Researcher from German Research Center for Artificial Intelligence (Towards Digital Twins of the Oceans: The Potential of Machine Learning for Monitoring the Impacts of Offshore Wind Farms on Marine Life
29 May 2023, Journal of Engineering
Better home management
Customers could be automatically switched to an energy tariff matched to them by artificial intelligence (AI), under plans being developed by Lightsource BP.
Powerverse, a startup created by the global solar developer, is launching a new home energy management platform aimed at alleviating growing “energy anxiety” among domestic consumers.
The firm’s chief executive told Utility Week it is aiming to attract 250,000 users over the next five years. He said that while the platform is not yet able to automatically switch customers between energy tariffs, the company is planning to add this feature in a future update.
Initially the tool will help users monitor and control their energy consumption and provide tailored advice on energy tariffs and home improvements.
The company said the adoption of low-carbon technologies such as electric vehicle (EV) chargers, solar panels, heat pumps and smart appliances is “time-intensive, confusing, and highly inconvenient, often requiring hours of research and causing ‘energy anxiety’. This is hampering the consumer-driven shift to a low-carbon, low-cost energy future.”
Powerverse said its platform uses AI fed with thousands of data points to analyse customers’ consumption patterns and provide tailored advice on actions they can take to save money or reduce their emissions, for example, by switching energy tariff or installing solar panels.
In the case of the latter, the company said the platform can use online maps, weather forecasts, consumption data and pricing from approved installers to determine the size and aspect of a customer’s roof, suggest suitable hardware, calculate the costs and benefits, and even arrange the installation itself.
The platform can also directly control devices such as solar invertors and EV chargers to enable customers to earn money by providing demand-side response.
Lightsource BP launches AI home energy assistant
24 May 2023, Utility Week Online
Analysis of a power grid in transition
Researchers at the University of Michigan will partner with a power grid technology company to study how electric vehicle driving and charging behavior impacts the electric grid.
The U-M Transportation Research Institute will partner with Utilidata to better understand how the increasing use of EVs can be expected to affect the electrical grid. The research will involve the use of artificial intelligence-powered technology, a first-of-its-kind platform.
Researchers have installed the technology on multiple EV charging stations across the U-M campus to collect data on grid impacts.
“As more people invest in electric vehicles, our electric grid needs to be ready to support the influx in energy demand,” said Josh Brumberger, Utilidata’s top executive officer.
“Access to real-time insights of when EVs are charging will help utilities identify charging locations and design better EV programs for customers.”
The company will use its “smart grid chips” to collect real-time voltage, current, and power data at the edge of the grid, which will allow researchers to analyze and find EV charging patterns at each location. The data will be studied alongside vehicle data from a group of participants in the project who have a vehicle-monitoring device installed on their EV.
Participants will have their start and stop time for charging, charging site, trips taken, and both acceleration and deceleration.
The idea is that closely analyzing driving and charging behavior will lead to a better understanding of how to manage EV demand on the grid, and help utilities develop customer charging programs.
Results of the research are expected later this year.
EVs are projected to comprise nearly 50% of all car sales by 2030 and Michigan’s goal is to build out the infrastructure needed to support 2 million EVs on its roads and highways by that time.
Impacts of EV charging on electrical grid focus of new AI study
23 May 2023, Jackson Citizen Patriot: Web Edition
Predicting grid instability
In a first phase, the research team ‘trained’ the algorithm on data relating to failures that occurred between 2015 and 2020 in a large electricity grid in the South, weather conditions (ambient temperature and humidity) and energy flows. with the aim of identifying possible correlations. In the following operational phase , the researchers tested the trained system for the analysis of a series of input data (not seen in the training phase). Among the algorithms tested, one in particular gave the most accurate results in terms of predicting future failures of the electricity grid studied as a function of both meteorological conditions and energy needs.
In the city, the grid is subject to greater load stress, due to the increase in the demand for electricity, concentrated in particular in the hottest hours of the day, due to the greater use of air conditioning systems. During the day, in fact, the air temperature often exceeds 40 °C and even during the night hours it remains above the historical average. “Our studies have shown that most of the failures occurred at the level of the cable joints and that, therefore, these elements suffer most from heat wave problems. This result provides a useful element to operators and manufacturers of electronic components, which will thus be able to conduct more targeted analyzes to obtain more resilient networks”, adds Valenti.
Energy: ENEA, artificial intelligence to avoid electricity blackouts from heat waves Distributed by Contify.com
18 May 2023, Contify Energy News
Assessing Solar Installations
For this, Microsoft Azure OpenAI Service will be used to help customers in the process of decision and contracting of solar panels through a self-developed chat based on GPT language models. Thus, a client will be able to easily consult any doubt related to the installation of photovoltaic panels, such as, for example, the ideal orientation of the same, the hours of sunshine that are in their location or the available subsidies.
Trend; Sapiens joins forces with Microsoft to leverage Generative AI for Insurers 333 words
28 May 2023, Insurance Newslink
NVIDIA Inception member Masterful AI has developed machine learning tools that can detect climate risks from satellite and drone feeds. The model has been used to identify rusted transformers that could spark a wildfire and improve damage assessments after hurricanes.
San Francisco-based Inception startup Orbital Sidekick operates satellites that collect hyperspectral intelligence – information from across the electromagnetic spectrum. Its NVIDIA Jetson- powered AI solution can detect hydrocarbon or gas leaks from this data, helping reduce the risk of leaks becoming serious crises.
NVIDIA -On Earth Day, 5 Ways AI, Accelerated Computing Are Protecting the Planet
24 April 2023, ENP Newswire
Industrial energy management
Sheetz, a leading convenience store chain, announced today that it will manage its energy use through Pear.ai, a state-of-the-art energy intelligence platform offered to customers by
Constellation and its affiliate company, Exelon Generation Services.
The Pear.ai platform provides businesses with utility expense management and centralized, streamlined access to all of their utility data as well as meaningful analytics. Sheetz will use Pear.ai to manage its comprehensive utility footprint, which includes power, gas and water across all 617 of its store locations.
“As a steadily growing business rooted in data and analytics, Sheetz is well positioned to take full advantage of the depth and flexibility of the Pear.ai platform and its supporting team,” said Mark Huston, president, Constellation’s National Retail Business. “We’re excited Sheetz will be able to leverage this solution to optimize its utility portfolio and manage its energy spend.”
The Pear.ai technology processes thousands of bills per week to identify bill anomalies, generate insights, and model predictive behavior through machine learning. Pear.ai also sends customer alerts to correct issues proactively using a direct conversational functionality that eliminates the need to wait, or pay additional fees, for information from an account or support representative. The technology has created meaningful savings for industrial, healthcare, retail store and higher education customers, among others.
Constellation: Sheetz Selects Pear.ai to Manage, Optimize Utility Expenses Distributed by Contify.com
9 February 2021, Contify Energy News
We have over 60 AI-enabled projects underway. Check out these 5 examples that show how AI R&D is utilizing and protecting our Nation’s vast fossil energy resources:
1. Smart Robots that Inspect and Repair Power Plant Boilers
Power plant boilers are the most important part of a power plant, but they are difficult and time-consuming for human operators to inspect and repair. To reduce risks and shorten maintenance and unplanned outages, FE is developing AI-enabled robots that can perform real-time, non-destructive inspection of boiler furnace walls. If they find a crack, they can operate repair devices to make an immediate repair, while using AI to enable smart data analysis and autonomy.
2. Drone-Mounted, Smart Methane Emissions Detection Systems
Methane emissions are of increasing concern to the oil and natural gas industry. They represent lost product and are a recognized greenhouse gas. To help reduce these emissions, a smart methane emissions detection system is being developed. It will detect methane leaks by pairing passive optical sensing data with AI algorithms. The system will also be mounted onto a drone, which will enable a more precise leak detection method.
3. Models that Predict Oil and Gas Well Productivity After Hydraulic Fracturing
The ability to design a more effective hydraulic fracturing program for a particular basin and predict how much oil and gas can be extracted from it is vital. However, executing a better design and obtaining accurate predictions is difficult because the process occurs underground. That is why AI is being used to make sense of multiple data sets and predict a well’s performance prior to drilling. These models will help energy producers optimize oil and gas stimulation and production and reduce environmental waste.
4. SMART-CS Initiative
Today’s technology can securely store captured carbon dioxide deep in the subsurface of the ground, but slow data processing can result in operational inefficiencies. To meet this challenge, FE developed a Science-Informed Machine Learning to Accelerate Real Time Decisions (SMART-CS) initiative. Using science-based machine learning and AI, this initiative will enable better reservoir management through more rapid decision making. It will develop real-time visualization, forecasting capabilities, and virtual learning environments. As a result, the SMART-CS initiative will help stakeholders and regulators overcome costly inefficiencies while increasing their confidence that the geologic carbon storage is secure.
5. Computers Dedicated to Fossil Energy Research
The Joule 2.0 supercomputer and the WATT computer, both housed at the National Energy Technology Laboratory (NETL), help accelerate the development of innovative, cost-effective technologies to ensure affordable, reliable energy for all Americans.
* Joule 2.0 supercomputer
Joule 2.0 allows researchers to model energy technologies, simulate challenging phenomena, and solve sophisticated problems using AI and other computational tools. A talented mathematician working 40 hours a week for 50 weeks per year would take about 55.9 billion years to do what Joule 2.0 can do in one second.
MIL-OSI USA: Using Artificial Intelligence in Fossil Energy R&D
10 April 2020, ForeignAffairs.co.nz
An Overall Summary
Here’s a summary of some of the key areas where AI has been utilized in the energy industry:
- Grid Optimization: AI algorithms are used to optimize the distribution of electricity across the grid, predicting demand and supply fluctuations, improving reliability, and reducing costs. By managing the grid more efficiently, energy waste can be minimized.
- Renewable Energy Integration: The intermittent nature of renewable energy sources like wind and solar makes their integration into the energy grid challenging. AI has been instrumental in predicting production from these sources based on weather forecasts and other factors. This helps to balance the energy supply from these renewable sources with the demand and other energy sources.
- Fault Detection: AI helps in identifying faults and potential issues within the energy grid in real-time. This allows operators to react quickly, minimizing downtime and improving the overall resilience of the grid.
- Demand Response: AI enables more advanced demand response strategies. By predicting energy usage patterns, utilities can incentivize customers to reduce their consumption during peak demand periods.
- Predictive Maintenance: AI can be used to predict equipment failures before they occur, allowing utilities to perform maintenance and avoid costly, unexpected downtime.
- Customer Engagement: Utilities use AI to provide personalized insights to customers about their energy usage, help them manage their energy consumption, and provide more responsive customer service.
Home Energy & HVAC Technologies
- Energy Consumption Prediction: AI algorithms can analyze historical energy usage data from smart meters to predict future consumption patterns. This can help individuals better manage their energy consumption and reduce costs.
- Smart Thermostats: AI-powered smart thermostats can learn from users’ habits and adjust heating and cooling systems to optimize energy use. They can also take into account external factors, such as weather conditions, to further optimize energy consumption.
- Home Energy Management Systems: These systems use AI to optimize the energy consumption of all the appliances in a household. They can manage when and how appliances are used based on the user’s preferences, real-time energy costs, and other factors.
- Predictive Maintenance: Similar to utilities, AI can be used in homes to predict when HVAC or other systems may require maintenance, thereby avoiding unexpected failures and improving overall energy efficiency.
On the other hand….
There is a downside to AI :
“At extreme scales, training the GPT-3 model just once consumes 1,287 MWh, which is enough to supply an average U.S. household for 120 years,” said Mosharaf Chowdhury, an associate professor of electrical engineering and computer science.