2022-089 – Local Adaptive Modular Protection for Electric Power Grids


The protection system is a crucial part of the electric grid for fast detection and isolation of faults. The conventional protection system lacks the intelligence required to modify its actions based on prevailing system conditions. To address this problem, adaptive protection was introduced. An adaptive protection system (APS) is a real-time system that can modify protective actions according to the changes in the system’s condition; however, APS depends significantly on the communication infrastructure to monitor the latest status of the electric power grid in order to send appropriate settings to all of the protection relays in the grid. Consequently, an APS is highly vulnerable to communication system failures. Thus, there exists a present market need for a solution that allows APS to continue to operate even if the communication system fails.

Technology Description

Researchers at the University of New Mexico and Sandia National Laboratories provide an approach of adding local adaptive modular protection (LAMP) units to the adaptive protective system (APS). Each LAMP unit is installed in parallel with the conventional protection relay used in APS. This addition of LAMP units to the protection system can guarantee its reliable operation under extreme events. LAMP units operating in parallel with the conventional APS will be able to act as a backup if APS fails to operate due to a communication system issue. LAMP units can also accommodate reliable fault detection and location on behalf of the protection relay. This proposed approach utilizes support vector machines which is a classification that classifies inputs with very high accuracy as a machine learning algorithm. The advantage of this approach is to accommodate a communication-free and setting-less protection that can adjust its reach in real-time based on the prevailing circuit conditions.


  • Offers more reliable protection system operation without any communication infrastructure requirement
  • Detects and identifies fault types
  • Fully setting less which can eliminate human errors
  • Identifies fault zones

Potential Applications

  • Adaptive Protection System
  • Electrical Grid
  • Machine Learning

Contact Information

Name: Andrew Roerick

Email: aroerick@innovations.unm.edu

Phone: 505-277-0608