[UC30] Suspicious Patterns Identification & Analysis

The Utility Problem

As energy is consumed daily, the utility experiences losses, classified as technical or non-technical. The total losses a utility experiences are simple to determine: subtract the energy bill from the energy supplied or input into the substation depending on the reach of the utility. However, differentiating between technical and non-technical losses and identifying the location are much more complex tasks, requiring analytics to be done efficiently. While they can be improved by updating and maintaining grid components to increase efficiency, technical losses are inherent in the grid and often of much smaller magnitude. However, non-technical losses, namely energy theft and failures (accidental or intended) of energy sensors, represent a much more significant loss for the utility and can be eradicated through analytics. 

Non-technical losses, as mentioned above, may occur for multiple reasons. Some examples are as follows: 

• An individual tampers with a power supply line to bypass the energy meter;
• An individual manipulates the energy meter to provide readings of less or no energy, 
• An individual accidentally causes the energy meter to provide inaccurate readings;
• Or there is a technical malfunction of the energy meter.

Thus, it is in the utility’s interest to identify and eradicate these losses. This effort is considerably eased with smart meter data (if available) as utilities gain valuable insights through the interval readings provided, enabling more in-depth load profiles. Without data, this effort is very manual and inefficient. Still, with data, the utility can identify suspicious patterns in the energy consumption profiles and investigate them further to determine these non-technical losses. 

Awesense’s use cases are constantly evolving. Request free access to our Sandbox trial and GitHub repository to see the full use case descriptions and implementations, and to learn how you can create new use cases yourself.