For LVDC network protection, the fault current raises much faster than that in AC, resulting in protection must operate with very high speed, commonly believed several ms. However, if connecting a inductor to constrain the change of rate of current, the fault current will be less "steep", causing long protection time. The thermal energy generated may also be harmful to the network.
How to determine the suitable speed for DC protection? What's the relation between the time that DC network can bear and the feature of the fault current?
IDMT(Isqr.R) with instantenious setting, relay could solved the problem, with suitable selection of fault reactor used....
IDMT does work in AC to realize coordination. However, DC network requires faster protection, so DC protection usually applies instant operation. IDMT cannot be used for discrimination. The problem is how long is allowed to do the instant operation.
Hi Chunpeng,
Could you tell us a bit about exactly what is the DC side configuration like ?
-Sanjay
The issue with DC is that we have embedded power electronics, and they have poor or limited short circuit capabilities. Converters' controls are also sensitive to large voltage drops that can be caused by short circuits. Also the post fault transient voltage will be higher if the fault stays longer (higher energy will be stored in the line inductor during the fault).
So the DC protection systems need to isolate faults far faster than conventional protection. Their operating time has to be fast enough to: prevent a high transient and steady state DC currents from damaging equipment; prevent the main converters from losing control and tripping unnecessarily; and reduce the impact of post-fault power quality and stability issues.
For further information you can read my recent IEEE paper:
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7519032&newsearch=true&queryText=emhemed
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering