Grid interrupted: Differential relay communications in smart grid

In this age of electrification, a resilient grid means never having to say you’re sorry for an unexpected power outage. Differential protection relays are poised to make this dream a reality.

These days, everyone wants more electricity from the grid. Industries are switching from fossil fuels to electricity to power everything from production equipment and delivery trucks to straddle carriers, boilers, and furnaces. Many increasingly rely on AI data centers, which have an insatiable appetite for electricity. Consumers are trading gas-powered cars and trucks for electric vehicles and heat pumps. And utilities are feeling pressure to meet these new demands and deliver customers safe, high-quality, interruption-free power. 

The “interruption-free” part of this equation has long been the stuff of dreams. Innovative network communications technologies for differential protection relays are poised to make this dream a reality.

Differential relays demand uninterrupted communication

When problems occur on power lines, utilities count on differential relays to reliably detect and isolate faults. Each relay demands uninterrupted communication with a remote counterpart to provide these capabilities. 

Most differential relays and other grid applications run on robust SONET/SDH- and TDM-based communications networks that have proven their worth for decades. Using these networks, utilities have largely succeeded in protecting their primary grid assets, including transformers, and minimizing disruptions to electricity supply across the grid during fault events. But SONET/SDH and TDM networks are now obsolete. Vendors no longer support them.

Many utilities are migrating grid communications to modern Internet Protocol Multi-Protocol Label Switching (IP/MPLS) based wide area networks (WAN) that can boost service performance, flexibility, and scalability. 

Utilities use IP/MPLS to embrace more applications for digital grid operations. These applications will provide them with a higher level of automation and better oversight to meet the demands of electrification and other global trends, such as the shift to distributed renewable energy generation and the rapid rollout of new AI data centers.

Is IP/MPLS ready to support relay communications?

A modern IP/MPLS communication infrastructure is the essential foundation for the digital grid. But the million-dollar question is this: Does IP/MPLS have what it takes to meet the stringent demands of relay communications?

Protection relays are highly sensitive to network impairments and disruptions. They need the network to provide reliable communications channels to transmit and receive current data with low and consistent network delay, delay variation (also known as jitter), and delay asymmetry.

Relays also need the network to deliver the strongest possible resiliency. If relay communications stop, so do relay operations. In other words, the grid’s reliability is directly tied to the reliability of relay communications.

IEC’s 61850-90-12:2020 technical report, which defines the WAN engineering guidelines for power utility automation, highlights the critical need for resilient relay communications. It specifies that there can be no recovery delay—that’s zero milliseconds—if there is a failure in the network path that carries differential traffic between the relays at each end of a given power line. 

This requirement exceeds the 60-millisecond recovery time utilities can achieve with advanced SONET/SDH- and TDM-based 1+1 network redundancy mechanisms such as automatic protection switching (APS) and subnetwork connection protection (SNCP). But, while recovery in 60 milliseconds is fast, it still results in a data loss that could disrupt relay operations, putting grid safety and reliability at risk—and could put customers’ electrification plans on hold.

This means utilities can’t continue using the same old redundancy approaches when shifting differential relay communications to IP/MPLS. It’s time for a new approach to seamlessly redirect relay communications to a redundant backup path if the active path fails.

Hitless IP/MPLS redundancy 

The good news is that such solutions exist today. Utilities can ensure highly resilient differential protection communications over IP/MPLS with a redundancy mechanism that combines two router innovations: hitless active multipath pseudowire redundancy and asymmetrical delay compensation (ADC). 

This new mechanism isn’t like conventional 1+1 redundancy. It doesn’t aim to detect network faults first and then switch data traffic from the active path to a backup path as quickly as possible. Instead, it simultaneously replicates relay data traffic and transmits it over multiple active pseudowire paths across the IP/MPLS network. The receiving router picks the best copy of this data traffic from the set of active paths and sends it on to the relay. There is no switchover time.

Active multipath pseudowire redundancy is a big deal because it ensures hitless multi-fault resiliency and meets IEC 61850’s zero-delay requirement. The relays don’t detect the network failure and continue communicating without interruption, meaning customers won’t notice the failure either. 

The mechanism uses IP/MPLS traffic engineering to ensure each active path uses a separate physical route to provide multi-fault resiliency. If network links fail on multiple active paths, the routers still receive data from the remaining active paths. There is no impact on relay communications or protection application performance.

Relay operations also have stringent demands regarding delay asymmetry caused by network jitter. Delay asymmetry can cause a relay to falsely trip a circuit because its current measurements don’t match the correct measurements received from the relay at the other end of the line. 

ADC technology, a smart buffer manipulation algorithm in the IP/MPLS router, mitigates the impact of asymmetrical delay on the forward and reverse communications paths between protection relays. It works seamlessly with active multipath pseudowire redundancy. ADC will also neutralize delay asymmetry caused by a link failure. 

An industry-Validated Innovation

Rigorous tests have been conducted with utility and telecom industry partners to validate the new redundancy mechanism. These tests aimed to show that IP/MPLS communications networks can provide performance and resiliency to meet the stringent demands of TDM- and Ethernet-based differential protection services. 

For the tests, many scenarios were attempted to break the network paths that relays rely on for communications, from fiber cuts and congestion to node, link, port, card, and router failures. They focused on ensuring the active/active multipath redundancy mechanism provides:

• Highly reliable and available communication and time synchronization
• Minimal network delay
• Suitable traffic segregation and quality of service (QoS) measures
• Constant, symmetric network delay with minimum jitter

The test results proved that the redundancy mechanism meets all these requirements and provides superior resiliency for line differential protection communications. When a network fault triggers redundancy protection, relay operations can proceed without interruption, and power lines remain protected. 

Utilities no longer have to worry about whether network problems will leave lines unprotected or cause unexpected power outages. Customers can get on with the business of electrification, knowing the grid is ready to support their daily needs and biggest ambitions without disruption.

Originally published on eepower.com and authored by Dominique Verhulst, Global Energy Practice Leader at Nokia