The energy sector is at an inflection point. Electric and gas utilities are facing a perfect storm of converging challenges that is forcing them to re-think how they manage their operations, re-invent their infrastructure and re-envision how they work with residential, commercial and industrial customers. These challenges include increased volatility of both energy supply and demand, aging infrastructure, the graying of the industry’s workforce, climate change, regulatory pressures, supply chain disruptions and rapidly evolving technologies.
Those challenges are significant, but utilities can successfully navigate them by having a well-defined AI strategy that puts Artificial Intelligence (AI) and Machine Learning (ML) to work in the right way. One of the pillars of every utility’s IT strategy should be to deploy AI/ML in conjunction with location intelligence — a powerful combination of technologies that can solve a long list of operational challenges.
Location intelligence refers to the actionable insights derived from the massive ocean of geographic data that organizations have at their disposal. Utilities are investing billions of dollars in connected equipment, smart machines, mobile tablets, sensors and IoT deployments that collect large amounts of data. But deriving insights from that data has been challenging. By combining location intelligence tools with machine learning algorithms in AI, utilities can automate analysis at scale to solve complex issues.
A comprehensive list of use cases for this powerful combination of AI/ML and location intelligence is too long to cover in a single article, but here are seven areas that utilities should target with their AI strategy:
1. Asset Management
Utilities can employ AI capabilities (e.g., image recognition) to pinpoint exactly where in the grid to perform asset inspections, maintenance, repairs and replacements. For example, a utility might capture multiple images of a single distribution pole (via remote aerial vehicles or ground crews). AI can ingest and process the images to detect anomalies such as a cracked insulator or missing pole cap. The AI solution can identify high-risk poles for the engineers to validate and dispatch field crews.
2. Decarbonization
ML solutions can learn from data patterns across customers, distribution infrastructure, and power generation assets, generating insights otherwise unavailable. This newly generated information can help utilities build infrastructure that meets net-zero requirements and goals. For example, geospatial-powered AI and deep learning models can identify suitable sites for solar panels. These tools can measure land elevation, evaluate land use data, assess weather patterns, incorporate environmental factors and determine proximity to existing infrastructure and grid assets — all in consideration of solar power generation potential.
3. Vegetation Management
Utilities can use imagery data and ML models to identify vegetation species and other attributes to prioritize actions. For example, tree species identification can help accurately determine when and where to perform tree trimming and vegetation removal. Utilities can employ AI to take advantage of these higher-quality datasets to detect unstable slopes and predict if there is a risk of landslides or if vegetation will slide into a transmission or distribution line.
4. Risk Reduction (Fire/Weather)
Predictive AI models can sift through historical and forecasted meteorological data to identify areas in danger of extreme weather or climate risks, helping utilities understand where to harden assets to protect vulnerable populations from hurricanes, flooding and wildfire events. Utilities can overlay asset, weather and location-accurate maps and perform ML-powered simulations around storm scenarios. They can process imagery to detect dead and diseased trees, determine fuel removal areas, and prioritize utility poles to harden and coat with fire-retardant material.
5. Grid Modernization
AI provides a foundation for grid modernization by accessing and analyzing vast volumes of real-time operational technology (OT) sensor data combined with IT application data. ML algorithms can then predict when grid components will fail and recommend when, where, and in what sequence to repair or replace parts. AI can also adjust power distribution based on analysis of demand patterns to optimize energy usage and lower costs. Real-time AI-enabled grid monitoring provides on-the-fly response capabilities to enhance capacity and reliability while reducing outages and mitigating their impact.
6. New Business Opportunities
Advanced Metering Infrastructure (AMI) 2.0 — the next generation of smart meters — offers significantly more data points to evaluate and drive decisions. Utilities can use AI to extrapolate and understand more details about residential and industrial power usage. For example, new smart meters can determine appliance-level consumption, which utilities can leverage to offer new products and services, such as upgrade rebates. By leveraging AI/ML, utilities can quickly combine and process diverse datasets and identify areas of opportunity to market renewable energy products and services.
7. Customer Service
Chatbots can use large language models and prompt engineering to guide customers through common complaints and topics. Based on predefined keywords and phrases and using location-based data and analytics, the chatbot will supply a robust, rich experience for users with service questions. Whether calling to report an outage or asking about the duration of a service disruption, customers can ask questions in their own words and receive a response with more helpful context and detail.
Here, Today, Now
I should take a moment to underscore that these use cases are not things to keep in mind for the future. These are applications of AI and location intelligence that can be put into action now to immediately make a positive impact. Utilities have already invested in generating and aggregating this operational data. By deploying the right AI/ML + location intelligence strategy, utilities can quickly transform that wealth of data into improved decision-making, optimized operations, reduced risk, improved customer service and increased safety.
