Protection and Control Redundancy Considerations in Medium Voltage Distribution Systems

By Mohammad Vaziri, PG&E, with Bogdan Kasztenny and Rich

Extra high voltage transmission and sub-transmission backbones provide for multiple power flow paths to increase availability of power supply and to facilitate work on the primary equipment while maintaining service. Redundant secondary systems are deployed not only for dependability but also to maintain the primary equipment in service during failures or planned outages of the protection and control schemes.

Distribution networks remain the last link between bulk loads and the power grid. As such, their reliability impacts the overall level of satisfaction with the power supply. In the age of deregulation and increased visibility into the power provider, the reliability of supply is measured through a variety of indices, such as average interruption duration index and average interruption frequency index.

The reliability solutions known and widely deployed in the high voltage transmission system are not used in the distribution network because of the cost implications. Building redundant power flow paths by adding feeders and transformers remains impractical; however, making extra investments in the secondary protection and control systems can be considered given the available technology. It is beneficial to revisit the basic principles beyond distribution networks created decades ago, in search for increased reliability given today’s cost/benefit opportunities.

Pacific Gas and Electric (PG&E) has taken action to increase the reliability of distribution networks via redundant protection and control systems. This redundancy can be accomplished on a very cost-efficient basis using modern protection and control solutions.

A Little Background: Redundant vs. Back-up Protection

The traditional method of maintaining reliability of the distribution system is to use time-coordinated backup protection as shown in Figure 1 (pg. 52). With this method, five relays are required for a four feeder distribution system (four primary protection relays and one back-up relay). The transformer overcurrent relay is the backup relay for all four feeders.

Click here to enlarge image

If the protection of one of the feeders is unavailable, the backup relay will operate for a fault on this feeder, and will isolate a larger part of the distribution system than just the faulted feeder. While less expensive to implement than a redundant scheme, back-up protection is usually slower and less selective than the primary protection.

In contrast, a redundant protection scheme is a method that maintains the correct operation of the distribution protection using redundancy of elements to eliminate single points of failure. This is shown in Figure 2 (below). With this method, eight relays are required for a four feeder distribution system (four primary protection relays and four redundant relays). All four feeders have fully redundant protection. As long as one of the two relays that are protecting Feeder 1 (Relay 1 or Relay R1) is operating, the protection for Feeder 1 is available. The principle can be carried forward and extended on relay input sources, independent routing on control wiring, redundant trip coils, and redundant batteries. These principles are followed in the extra high voltage installations. Distribution networks can focus on major components of the secondary system given the actual field record for reliability of various components. In this respect, redundant instrument transformers, trip coils or batteries are not necessarily required for major improvements in the overall reliability. The exact extent of carrying the principle of redundancy needs to be driven by a cost/benefit ratio. For example, medium voltage CTs are relatively inexpensive and can be considered, while dual trip coil medium voltage breakers are hardly available.

Click here to enlarge image


Protection Redundancy Implementation Methods

Protection functions are made redundant by simply adding more relays for the primary zone of protection. Three methods for redundant relaying include dual-redundant relay protection, relay pair protection, and multiple-source relay protection.

These schemes must be carefully implemented to prevent mis-operations from occurring during both in-service and maintenance conditions. There are several methods available for supplying redundant protection, depending on the relays selected for use, the need for additional functions in the relay, and the ease of implementation.

Dual redundant relay protection: Dual-redundant relay protection uses two feeder relays for each feeder circuit. This method provides complete redundancy of short circuit protection as shown in Figure 3 (above), and can provide complete redundancy of control functions, metering, and communications, depending on the specific implementation. One typical implementation is to use a full-featured feeder management relay that includes protection, metering and control functionality in combination with a less expensive feeder relay that provides only short circuit protection. Another option is to use two feeder management relays that have similar capabilities in protection, metering, and control, in a Set A / Set B combination.

Click here to enlarge image

Relay Pair Protection: Dual redundant relays provide completely redundant protection, but they can be expensive. Some modern microprocessor relays have multiple sets of three-phase and ground current inputs, with independent overcurrent protection for each set of current inputs. This allows one relay to be the primary protection for one feeder, and the redundant protection for a second feeder, as shown in Figure 4 (page 53). Therefore, with feeder relay pairs, two relays can protect two feeders with complete redundancy, for the cost of one standard protection package.

Click here to enlarge image

Multiple-Source Relay Protection: Some microprocessor-based relays can accept up to six separate three-phase and ground current inputs, and provide independent overcurrent protection for each of these inputs. This is a very cost-effective method to add redundant overcurrent protection, as one additional relay can provide redundant overcurrent protection for a small distribution substation or switchgear lineup with up to six primary protection relays. This method is illustrated in Figure 5 (below).

Click here to enlarge image


PG&E’s Redundancy Objectives

Pacific Gas & Electric recently developed the Integrated Protection and Control (IPAC) standard protection scheme for their medium voltage distribution feeders. The IPAC system is delivering a significant operational benefit for PG&E, including reducing capital, maintenance, and operating costs, increasing the information available from a substation, and more tightly integrating Supervisory Control and Data Acquisition (SCADA). One of the business and technological goals of the IPAC system is improving system reliability while decreasing the service downtime for greater customer satisfaction.

PG&E adapted principles from the North American Electric Reliability Corporation (NERC) and Western Electricity Coordinating Council (WECC) criteria for protection of bulk lines to distribution networks when designing the IPAC system. Specifically, the IPAC system uses redundant feeder protection relays, from different manufacturers, to provide protection and control functions for a distribution feeder.

The IPAC system uses a dual-redundant scheme, implemented in two feeder management relays. PG&E feels this method provides the greatest benefits in availability of the protection system, operational flexibility, and testing and maintenance.

All of the basic protection functions are implemented in both the Set A and Set B relay, including directional control of overcurrent functions, undervoltage protection, and overvoltage protection. The IPAC system also uses independent sets of CTs for the Set A and Set B relays. This increases the overall availability and reliability of the system, for the cost of inexpensive medium-voltage-rated CTs.

To simplify the design, and in keeping with the goal of eliminating or limiting the impact of a single point of failure, PG&E splits control functions between the Set A and Set B relays. Most of the control functions, including reclosing, breaker failure, underfrequency load shedding and local control operations, are provided in the Set A relay. The Set B relay is responsible for SCADA communications, and remote control of the distribution feeder.

The split of local control operations and remote control operations between the Set A and Set B relay is intended to provide demarcation between local and remote control of the feeder. This simplifies the scheme for operations personnel, and simplicity helps maintain reliability. Splitting control between the two relays complicates the design and engineering of the original system, and requires substantial contact input/contact output communications wiring between the two relays. This integrated control has been carefully tested to ensure successful operation of the feeder.

The operating states of control functions, such as neutral overcurrent blocking, reclosing sequence, and active settings group, is determined by local or remote command. This command must be sent between the Set A and Set B relays to achieve coordination of the control function. When one relay is returned to service after maintenance, the operating state of all control functions must be synchronized to the operating state of the operating, “in service” relay.

PG&E designed their IPAC system to take full advantage of the benefits of microprocessor relays, using the dual redundant protection and control method as their new standard for medium voltage distribution feeders. Specifically, the IPAC system provides:

  • Completely redundant protection functions for short circuit and voltage-based protection,
  • Clear demarcation between local and remote control of the distribution feeder, and
  • Integration between the Set A and Set B relay to properly execute control functions and synchronize settings.

Mohammad Vaziri is supervising protection engineer at PG&E.

Bogdan Kasztenny is manager, protection & systems engineering, at GE Multilin.

Rich Hunt is application engineer at GE Multilin.

Previous articleELP Volume 85 Issue 6
Next articleHow advanced metering impacts a call center
The Clarion Energy Content Team is made up of editors from various publications, including POWERGRID International, Power Engineering, Renewable Energy World, Hydro Review, Smart Energy International, and Power Engineering International. Contact the content lead for this publication at

No posts to display