By Matthew H. Tackett, P.E., Tackett Electric Company Inc.
Effective short-circuit and instability protection systems can do much to prevent cascading outages like the one that occurred in the Northeast on Aug. 14. However, a number of changes are needed in protection system philosophy and design to minimize the probability of future cascading events.
Purpose of Transmission System Protection
There are only two reasons a transmission facility should be automatically removed from service by a protection system. One is to quickly isolate a short-circuit fault, and the other is to isolate a small unstable area of the grid. The automatic removal of a transmission facility for any other reason will compromise overall grid stability and increase the probability of massive cascading outages.
The Need for Short-circuit Protection Security
Short-circuit protection security refers to the ability of a protection scheme to avoid unnecessary operation. The unnecessary operation of a short-circuit protection scheme compromises system reliability by introducing additional transient disturbances, weakening the system and creating an environment that promotes uncontrolled cascading outages. Short-circuit protection security is achieved by utilizing short-circuit transmission protection schemes that respond only to short-circuit faults.
Short-circuit protection systems should not respond to overloads. While overloads can result in facility damage and/or conductor clearance violations, it is better to allow system operators to address loading problems via manual system adjustments (e.g., redispatch, etc.) than to risk a cascading outage of many transmission lines due to overload tripping.
Nor should short-circuit protection systems respond to power swings. While power swings may indicate a potential stability problem, it is better to address the stability problem via a coordinated system-wide automatic instability protection system that isolates unstable areas in a controlled manner than to risk a massive uncontrolled cascading outage that results from the simultaneous or near simultaneous response of many short-circuit protection schemes to power swings.
The Need for Short-circuit Protection Dependability
Short-circuit protection dependability refers to the ability of a protection system to operate when necessary. Dependable protection systems are absolutely necessary to prevent short-circuit faults from creating transient stability problems. However, to ensure dependability at minimal cost, protection system security has often been neglected. If the overall reliability of the interconnected grids is to be improved, protection system dependability must be maintained, but in a manner that does not compromise protection system security.
The Need for More Effective Instability Protection
While out-of-step protection schemes have been employed on some transmission facilities in the past, these protection systems are not always effective. For one thing, these schemes are not always capable of isolating unstable areas in a manner that preserves balance between generation and load. In addition, while these schemes can detect power swings, detection is often based on impedance measurement rather than voltage measurements, and this limits the flexibility of these schemes. Furthermore, protection schemes are needed to address voltage instability as well.
The industry should move toward the replacement of traditional short-circuit protection schemes, such as carrier blocking directional comparison schemes, with redundant numerical line differential schemes that utilize redundant fiber optic pilot wire channels. This will eliminate any response of the protection system to overloads and/or power swings. That is, the short-circuit protection system will be designed to respond only to short-circuit faults. In addition, unlike many of the protection schemes employed today, backup protection is provided via scheme redundancy–thus the failure of a protection scheme component will not result in a corresponding increase in fault clearing time and/or the number of facilities tripped (breaker failure is an obvious exception to this). While redundant differential pilot wire schemes will represent considerable investment, the return on this investment could be significant if future widespread outages can be avoided.
To mitigate the impact of system instability, a coordinated system of instability protection schemes must be used. While out-of-step protection schemes have been utilized in the past, their effectiveness has been hampered by the inability to preserve balance between generation and load following the isolation of an unstable area.
This problem can be addressed via the dynamic supervision of instability protection schemes from a central grid computer. That is, instability protection schemes would be installed on all transmission lines, but only a handful would be enabled by the central grid computer at any given time. The central grid computer would partition the entire interconnected grid into relatively small zones of balanced generation and load on a periodic basis (perhaps every 30 minutes) based on real-time SCADA data and the results of coordinated EMS state estimator runs.
The grid computer would also attempt to establish these zones such that reactive power resources (static and dynamic) are balanced with reactive load. Once the zones are established, the central grid computer would enable instability protection schemes only on lines or transformers that interconnect two different zones. Should one or more zones become unstable, instability protection schemes would isolate these unstable zones, allowing the rest of the grid to remain in balance. Instability protection schemes on all other facilities would be disabled, thus increasing the chances for recovery following a severe transient disturbance.
Given that communications channels are required to support the numerical differential protection schemes, these communications channels could also be utilized to facilitate angular instability detection schemes that compare the sending-end and receiving-end voltage angles on a continuous basis. If engaged, these schemes would isolate areas when the angular separation between voltages exceeds a predetermined level for a predetermined duration. The rate of change of angular separation could also be used as an input for the instability protection scheme.
In addition to angular instability protection schemes, voltage collapse detection schemes should be established. While there are a number of ways in which voltage collapse detection schemes could be designed, isolation of the voltage collapse would be made via lockout of only those lines that interconnect the collapsed zone with other zones.
Improved transmission protection systems can assist in preventing massive cascading outages. However, to minimize the frequency of smaller cascading outages that result when unstable zones are isolated, transmission expansion and solid operating policies will be necessary as well.
Matthew H. Tackett is president of Tackett Electric Company, an electric utility consulting and software firm specializing in power delivery reliability and electric utility deregulation. He holds a B.S.E.E. degree from Purdue University and is a registered professional engineer.