How can utilities use smart metering to better understand energy usage patterns?

On the cynical side of that topic, you also have to remember the utility can analyze the data to optimize where peak demand is used to their benefit to potentially increase pricing to make up for the reduction in energy for conservation. The utilities job is to make money and as customers potential demands can go down that’s lost revenue, they will try to make up for it.

Having worked with companies in the great white North, weather has a significant impact on energy consumption. Heating and cooling demands increase during extreme temperatures, affecting both residential and commercial energy usage. Energy consumption tends to fluctuate with the seasons. For utilities, having insights into usage patterns helps in grid planning and management. This is crucial in places where negative and triple digit temperatures are the norm and customers rely on the grid for their own health and safety.

One key component missing from energy usage pattern analysis is the social and human elements to customer behaviour. The rebound effect is a well known example but more can be done to study how psychological and social elements can change consumer behaviour. For example, an apartment might have above average air-conditioning usage for noise reduction reasons instead of temperature. Thus for this case, a noise reduction window can be the solution instead of a higher energy efficiency system.

We have to be clear that smart is not more data or greater resolution. Smart means turning data into information into actionable and relevant knowledge for decision making. Thus, the information architecture of a system is equally as important as the hardware or software architecture.

There are several challenges companies are facing in regards to the energy usage patterns: 1. Latency of Data: Energy consumption data can be vast and complex, especially when collected in real-time data and the analytics is crucial. This requires a big enough server, smart analytical methods and techniques to process, analyze, and interpret. 2. Variability and Seasonality: Energy usage can vary greatly depending on the time of day, season, and weather conditions. This can be difficult to discern consistent patterns or predict future usage.

The greatest challenge is the translation of energy patterns as every case is different due to a million reasons from geography, season, industry and even changing user profiles. The state of art today is using the parento principle of focusing on the top 20% factors that affects 80% of the pattern. This approach has worked for the past decade in the early stages of energy efficiency understanding to find rather obvious opportunities but has a diminishing return year on year. The answer has to be the use of Model-based System Engineering (MBSE) tools and methodology powered by AI engines.

The field of digital engineering as advocated by INCOSE will help progress the art of energy usage understanding. Emerging system engineering tools like MBSE, Digital Twins and System Dynamics that can be enhanced on AI engines are helping utilities make significant progress.