What is AI? How does this help the Electric Grid Transformation?

Basic breakdown of AI components (Image Reference below)

Recently, I had the pleasure of attending several in-depth and thought-provoking courses on Artificial Intelligence (AI). I wanted to share with you all some of my personal learnings from those courses to bring you all along this journey with me.

Before jumping into how AI will help augment the existing workforce (and not replace folks) and in a way accelerate the transition to a sustainable grid for the 21st century and beyond, I would like to discuss some of the basics of AI that will help set the context for the discussion relevant for grid operations and the overall electric utility industry.

So, what do we mean by Artificial Intelligence? AI is a diverse field within computer science, aimed at developing systems that can perform tasks which typically require human intelligence. These tasks range from learning, reasoning, and problem-solving to perception, language comprehension, and potential self-correction. AI aims to emulate human cognition to execute tasks as varied as solving puzzles, understanding languages, and making decisions or predictions. (Spoiler alert: This definition was provided by ChatGPT! 😊)

AI is not a new concept in our lives; it manifests in web search engines like Google and Bing, speech recognition technologies like Siri, and facial recognition technologies such as FaceID. These applications employ various AI components like Natural Language Processing (NLP), Deep Learning, and Computer Vision. Even more complex examples include self-driving cars from companies like Waymo and Tesla, which integrate all these components along with Robotics.

So, then the obvious question, what is all the fuss about and what is new? The answer to that question is - Generative AI (GenAI). GenAI is a set of complex models and algorithms that can generate completely new data that is similar in structure and content to the data it has been trained on. Unlike the models mentioned above, which classify or differentiate between existing data points, generative models can create completely new data. These can range from text and images to more complex structures like code snippets, medical records, or even entire simulation environments. We have seen an explosion of GenAI applications that have been made available to the public that are more chatbot in nature, where a user interacts with these AI models by entering text (Prompt) that are able to then invoke the right response from the GenAI model that you as an end user is looking to generate. The larger application of this technology is going to be seen in large enterprises across sectors/industries and across job functions.

There is a lot more that can be covered on AI, it's components and other examples that I will talk about in future blogs but with this 30,000 feet overview of what AI is, we can jump into its application in the electric utility industry. In today's technological landscape, Machine Learning (ML) and other AI categories are already successfully employed in various areas of the electric utility industry. These include load flow analysis, state estimation of large electric power system models that are being operated and maintained by the transmission operators as well as the ISOs and RTOs, and leveraging smart meter data to predict customer energy usage for future billing purposes as well as for outage prediction to a certain degree in outage management systems (OMS) are just a couple of examples.

In the future, with advanced AI/ML models that will be leveraging GenAI, they will be leveraged in the following areas:

1. Predictive Maintenance: Generative AI can be utilized to develop predictive models that analyze data from various sensors placed on equipment to predict when maintenance is required. This approach not only reduces downtime but also extends the lifespan of the equipment, ensuring a more streamlined operation.

   
   

2. Demand Forecasting: By analyzing historical data and other relevant variables, generative AI can create more accurate demand forecasts. This assists utilities in efficiently managing their resources and ensuring a stable supply of electricity, adapting dynamically to changing demand patterns.

   
   

3. Customer Engagement: Utilities can leverage generative AI to create personalized customer engagement strategies. For instance, AI can analyze customer data to create personalized energy-saving tips or to develop new services that meet the individual needs of each customer, enhancing customer satisfaction and engagement.

   
   

4. Distribution Energy Resources (DER) Management: In the development and deployment of Virtual Power Plants (VPPs) into the distribution power grid, generative AI will play a crucial role by optimizing the distribution of energy resources (DERs). It can analyze data from various sources to devise strategies for the most efficient use of renewable energy sources, battery storage systems, and demand response strategies, fostering a more sustainable energy landscape.

   
   

5. Training Simulations: Generative AI can be used in the future to create realistic training simulations on the distribution power grid for operators in the control room on the fly without onerous configurations. Currently, these simulations are manually built by either software developers in the backend or are developed by a trainer who requires training himself to setup these simulations.

   
   

These examples only scratch the surface and in fact are being worked on by a number of software vendors actively today to augment their existing solution stack with AI/ML technology. Once the full potential of these models and algorithms are fully understood, there will be a tremendous number of use cases that will be identified and worked on in the upcoming years. In conclusion, understanding AI's full potential will reveal not just its technological prowess, but its potential to sculpt the future. As business leaders, tech enthusiasts, and consumers, gaining clarity on these components serves as a springboard for further exploration and discussion on leveraging AI for modernizing the utility industry.

For folks who have read this far and are interested in connecting with me on how I can help you in your journey, then feel free to reach out to me via email – abhinav@grid-transformation.com or via cell - +1-437-427-5155.

If you no longer wish to receive marketing emails or newsletters from Grid Transformation Inc., please email us atcontact@grid-transformation.com.

Image Reference - https://www.aspendigital.org/report/ai-101/