Principle of discrimination
Protection functions form a consistent system depending on the overall structure of the power distribution system and the neutral earthing arrangement. They should therefore be viewed as a system based on the principle of discrimination, which consists of isolating the faulty part of the power system and only that part as quickly as possible, leaving all the fault-free parts of the power system energized.
Various means can be used to implement discrimination in power system protection:
- Time-based discrimination,
- Current-based discrimination,
- Discrimination by data exchange, referred to as logic discrimination,
- Discrimination by the use of directional protection functions,
- Discrimination by the use of differential protection functions,
- Combined discrimination to ensure better overall performance (technical, economic), or back-up.
Principle of Time-based discrimination
Time-based discrimination consists of assigning different time delays to the overcurrent protection units distributed through the power system.
The closer the protection relay is to the source, the longer the time delay.
The fault shown in the diagram opposite (fig. 1) is detected by all the protection units (at A, B, C, and D). The contacts of delayed protection unit D close faster than those of protection unit C, which themselves close faster than those of protection unit B…
Once circuit breaker D tripped and the fault current has been cleared, protection units A, B and C, which are no longer required, return to the stand-by position. The difference in operation time DT between two successive protection units is the discrimination interval.
It takes into account (fig. 2):
- Breaking time Tc of the downstream circuit breaker, which includes the breaker response time and the arcing time,
- Time delay tolerances dT,
- Upstream protection unit overshoot time: tr,
- A safety margin m.
ΔT ≥ Tc + tr + 2dT + m
Considering present switchgear and relay performances, ΔT is assigned a value of 0.3 s.
Example Tc = 95 ms, dT = 25 ms, tr = 55 ms; for a 300 ms discrimination interval, the safety margin is 100 ms.
Advantages of Time-based Discrimination
- It provides its own back-up; for example if protection unit D fails, protection unit C is activated ΔT later,
- It is simple.
However, when there are a large number of cascading relays, since the protection unit located the furthest upstream has the longest time delay, the fault clearing time becomes prohibitive and incompatible with equipment short-circuit current withstand and external operating necessities (e.g. constraint imposed by utility).
This principle is used in radial power systems. (fig. 4). The time delays set for time-based discrimination are activated when the current exceeds the relay settings. The settings must be consistent.
There are two cases, according to the type of time delay used:
Definite time relays (fig. 5)
The conditions to be fulfilled are: IsA > IsB > IsC et TA > TB > TC.
The discrimination interval ΔT is conventionally in the range of 0.3 seconds.
IDMT relays (fig. 6)
If the thresholds are set to the rated current In, overload protection is ensured at the same time as short-circuit protection and setting consistency is guaranteed.
IsA = InA, lsB = InB, and IsC = InC
The time delays are set to obtain the discrimination interval DT for the maximum current seen by the downstream protection relay. The same family of curves is used to avoid overlapping in a portion of the domain.
Resource: Protection Guide 2003 – Schneider Electric