By Satish Saini, HEXstream utilities industry specialist 

Legacy AMI was developed to help utilities automate and efficiently manage customer meter readings and billing. But over the time, the availability of various key parameters (such as energy and power usage, voltage profile, current and meter-status data) have been utilized for broader applications beyond billing and customer services.

Real-time data of these key parameters, along with historical trends from MDMS or through external third-party vendors for data platforms using advanced data analytics, have proven to be quite valuable for utilities in efficiently and accurately managing various other functionalities and operations—think outage management, supply/demand balances through demand-side management/demand-response programs, asset management including asset-loading/overloading and performance insights for effective operational and maintenance decision-making.

With expanding applications of artificial intelligence and machine learning in AMI data analytics, modern utilities are transforming this data into advanced actionable intelligence for performance analytics. They are using this data for sharper predictive and condition-based maintenance of critical grid assets such as distribution transformers, feeders, breakers, line reclosers and more.

AI helps analyze historical trends, pattern recognition, and anomaly detection in key operational parameters. It also helps with optimizations and forecasting for improved asset health, system reliability, planning and grid efficiency using stored historical data and near real-time data availability.

Below, see examples of specific use cases in utility asset management:

• Predictions on assets failures using machine-learning models based on key operating parameters like voltage, current, harmonics, temperature and loading profiles. These predictions can prevent unplanned outages and extend asset life.
• Assets-performance analytics and insights (APA/API) using AI to calculate asset health indexes and/or risk indexes using various data platforms for asset loading, imbalance, losses, thermal stress, outage history, testing parameters/KPIs and maintenance records. This helps in prioritizing risk-based maintenance activities.
• Asset loading and overloading with deep-learning models to predict and issue alerts on overloads and other anomalies using historical patterns and available real-time data. This helps with proactive load balancing or shifting and informs smarter asset replacement and upgrade planning.
• Asset utilization and optimization using AI-developed load patterns of assets by timeframe and geographical locations to identify under/over-utilized assets. This helps with asset swapping, load shifting and optimizing capital investment.

*Find the previous installments in our AMI series:

Win/Win—AMI Analytics Empowering Utilities And The Customers They Serve

Benefitting Both Sides: AMI Analytics For Asset Management And Load 

WANT MORE? CLICK HERE TO CONNECT WITH US TO LEARN HOW YOUR AMI DATA CAN ENHANCE YOUR ASSET-MANAGMENT STRATEGIES.