Migrating Teleprotection Systems to Next-generation Networks

A teleprotection system communicates over a PSN in the event of a power transmission failure.

by Kobi Gol, RAD Data Communications

Teleprotection signals from protective relays are critical, even paramount, in this industry. They help manage power grid load and protect network equipment from severe damage.

By enabling load-sharing, grid adjustments and immediate fault clearance, teleprotection ensures an uninterrupted power supply. Protection commands, therefore, must be assured immediate delivery when problems are detected, enabling faulty equipment disconnection before systemwide damage occurs. As utilities move from legacy synchronous digital hierarchy/synchronous optical network (SDH/SONET) communication networks to packet-based networks, however, the complexity in ensuring protection performance is compounded.

Utility communications networks have been SDH- and SONET-based. That is changing as legacy infrastructure and substation devices migrate to Ethernet transport and Internet Protocol (IP)/packet-based networks. The smart grid evolution is a key driver for this change because packet transport offers high capacity at a lower operating cost. This will help handle traffic generated by advanced grid applications.

Next-generation supervisory control and data acquisition (SCADA) systems, wide-area situation awareness synchrophasor measurements and IP-based video surveillance are a few new applications that mandate the use of packet-switched networks. In addition, recent developments in substation automation, such as the International Electrotechnical Commission (IEC) 61850 standard, also require Ethernet capabilities throughout transmission and distribution.

Proceeding With Caution

Utility companies have been cautious about their IP transformations. They must maintain equivalent reliability to SDH/SONET systems while holding the line on capital expenses. Utilities also want to avoid overburdening operations as they introduce new devices and communications technologies, especially when migration includes the ongoing coexistence of SDH/SONET and next-generation networks.

Packet-based traffic management and hierarchical quality-of-service tools handle incoming data through classification, hierarchical scheduling, metering and policing of the traffic flow, shaping and transport over EVCs.

Implementing smart communications over packet-based networks requires reliable service assurance tools to ensure low end-to-end delay, the highest availability and network resiliency. For teleprotection, packet networks are incompatible with the requirements of extremely low symmetrical delay and minimal delay variation, also known as jitter. Ethernet technology has matured and can overcome these impairments and ensure appropriate performance, however. Typically, teleprotection signals carried over packet networks use a technology known as pseudowire emulation, which creates a tunnel for them through the packet network.

The key criteria for measuring teleprotection performance are:

  • Command transmission time, the overall operating time for a teleprotection system. This includes command initiation time at the transmitting end, propagation time over the communications link, and the selection and decision time at the receiving end, including any additional delay resulting from a noisy environment.
  • Dependability. This includes the assurance of issuing and receiving valid commands in the presence of signal interference or noise and minimizing the probability of a missing command.
  • Security. This includes the prevention of false tripping resulting from a noisy environment.
  • Bandwidth. This includes the rate used by the teleprotection system.
  • Resiliency. This includes the ability of the system to recover from failures.

The Challenge of Latency

Latency, or signal delay, requirements for utility networks vary, but most line equipment can withstand shortage or interruption faults of up to five power cycles. After that, the equipment might sustain irreversible damage or the fault might affect other network segments. In 50-hertz lines, this translates to total fault clearance time of 100 milliseconds. As a safety precaution, the actual operation time of protection systems is limited to 70-80 milliseconds, including fault recognition, command transmission and line breaker switching time. Some system components, such as large electromechanical switches, are less responsive and eat up most of the total time, leaving only a 10-milliseconds window for the communications element. New IEC standards are even more stringent regarding protection messages.

There are unavoidable sources of latency in a teleprotection system from the equipment itself through the signals’ conversion for transmission through the packet network. Delay can be addressed through traffic management tools. These ensure that teleprotection signals receive the highest transmission priority and minimize the number of jitter-inducing routing points traversed.

Advancements in Ethernet technology brought sophisticated mechanisms, which provide protection signals quality of service and priority. By properly managing bandwidth consumption and transmission priorities, predictable latency and jitter performance can be assured across the service path. Techniques involved include traffic classification, metering and policing, shaping, packet editing and marking, and hierarchical scheduling.

Assuring Smooth Operation

It is essential to test, monitor and troubleshoot the communications links to maintain smooth operations. Carrier Ethernet offers many tools for this: activity verification to stress testing; performance monitoring; and fault detection, propagation and isolation. Remote testing, end-to-end visibility and proactive monitoring capabilities help utility network operators anticipate service degradation. This allows them to reduce the number of truck rolls and on-site technician calls, assuring consistent performance and reduced operational costs.

Teleprotection performance is tested using a simulated power grid over Ethernet and MPLS.

Because packet communication networks are nondeterministic and were not designed with built-in synchronization mechanisms, certain complementary clock transfer solutions are required to ensure a stable network with predictable performance. This is particularly true when it comes to delay- and jitter-sensitive applications such as teleprotection and SCADA. Until recently, the typical, yet expensive, solution was to install a GPS at each node or service point.

Several methods can ensure synchronization in an all-packet environment. The two most popular are synchronous Ethernet (Sync-E), which uses the Ethernet physical layer to distribute frequency accurately, and precision time protocol (IEEE 1588), which involves a time stamp information exchange in a master-slave hierarchy to deliver clocking information. Adaptive clock recovery is another method, which distributes the clock over the network via pseudowire.

The sensitive nature of power protection applications requires much clock precision. IEC standard 61850 specifically addresses utility networks’ needs in timing and synchronization over packet networks, requiring accuracy on par with GPS levels. Thus, a utility must be sure its teleprotection network solution supports clock transfer, assuring GPS equivalency and significant cost savings over actual GPS installations.

Selecting a Solution

When migrating to next-generation packet networks, a utility’s technology decision depends on the number of sites to be connected, site size and the ability of the selected solution to ensure consistent performance across the different access media available at each site. Available options include multiprotocol label switching (MPLS) using virtual private LAN service (VPLS) encapsulation, IP/MPLS or Ethernet end-to-end, or a combination of Ethernet access and an IP/MPLS core. An end-to-end VPLS might provide the required resiliency for critical applications. It raises security issues, however, and offers limited performance-monitoring tools.

Traditional TDM traffic is encapsulated as pseudowire packets and tunneled through the Ethernet, IP, or MPLS network, then restored as TDM traffic.

A combination of Layer 2 Ethernet access with an IP/MPLS core offers a lower cost per port, more robust tools and advanced protection mechanisms via its switching infrastructure. In addition, it allows utility network operators to maintain their existing access media installed base, fitting well in distributed sites with copper, fiber and wireless infrastructure.

The smart grid evolution and migration to next-generation packet networks in utility communications are ongoing and inevitable. This evolution might create challenges when it comes to critical applications such as teleprotection, but as long as utilities’ network choices reflect the proper priorities, they can avoid trouble. Hybrid packet and legacy solutions that meet the exacting performance criteria of minimal transmission time, reliability and security are the most viable alternatives because of extremely low symmetrical delay, robust clock accuracy, quality of service assurance, resiliency and ongoing performance monitoring.

Kobi Gol is business development and solution manager for utilities, transportation and migration at RAD Data Communications.

More PowerGrid International Issue Articles
PowerGrid International Articles Archives
View Power Generation Articles on PennEnergy.com


Previous articleBuilding Brains Behind the Smart Grid
Next articleCreating a Smart Grid Road Map

No posts to display