Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Do Settingless Relays truly improve Protection Ideas?

  1. Apr 4, 2016 #1
    I have more of a discussion type question than one about Electrical Concepts (though I suppose it will devolve into that area). I've been doing a bit of reading about settingless relaying schemes and was wondering how they really improve over the traditional methods?

    The method of Dynamic State Estimation seems to do precisely the same thing as any other differential protection scheme, except it looks to be better at detecting high impedance faults.

    Simplicity is beautiful, and, in many areas, superior to the obfuscation that "new technology" tends to introduce. If protection schemes are meant be as simple as functionally possible, would it be entirely prudent to complicate the methods by introducing State estimation and communication between remote areas? Thoughts?
  2. jcsd
  3. Apr 4, 2016 #2


    Staff: Mentor

    The KISS principle versus "smart complexity" That's a pretty philosophical topic.

    I could absolutely imagine that a state estimator, if working correctly, could detect many abnormal conditions on the grid, or in a power plant, or in almost any engineered creation. The problem is the "if working correctly" qualifier.

    Need I point out that more complex software-based gadgets not only have more ways to go wrong naturally; they also necessarily make us more vulnerable to cyber attack.

    Any KISS versus complexity debate must reduce to costs versus benefits. Costs and benefits can best be estimated by, first stop drinking the EPRI cool aid, and second by installing pilot advisory only demonstrations for a number of years. Collect data on reliability, true and false positives and true and false negatives.

    Instead of the power grid, you might get a better conversation raising the same points in a Boeing versus Airbus philosophical debate about automation in aircraft cockpits.
  4. Apr 4, 2016 #3
    This corroborates my thoughts. There doesn't, at first glance, seem to be enough economic "skin in the game" to get all excited and change an entire paradigm of thought. HIF are a problem, but not one that occurs frequently enough and in a manner damaging enough to worry too much about. The other proposed areas of improvement are similarly "small." One could correctly correlate this to the idea of High volume, slim margin: Settingless relays may make sense, but only over several years, and assuming the "if working correctly" caveat is not violated significantly within that time frame.

    I'm not the one to speak intelligently on Cyber security as my knowledge of the discipline is lamentably slack, but the word "necessarily" caught my attention (as was, perhaps, intended). Would you explain?

    I'll have to look-up more about the Boeing vs Airbus debate before I dare tread those pastures. ;)
    Last edited: Apr 4, 2016
  5. Apr 5, 2016 #4
    Perhaps in this instance the simplest explanation is, if you don't need to run an ethernet cable to it (ie mechanical safetys, contacts ect) you can't exactly hack into it. In other words what's not on a network is by definition not open to network intrusion. Introducing a cyber element into anything necessarily opens this door as no standalone system can differentiate between auhorized and unauthorized access.
  6. Apr 5, 2016 #5


    Staff: Mentor

    Good advice. But the Stuxnet virus demonstrated conclusively that air-gapped systems with no net connection, can still be hacked.

    Even if the probability of a successful hack is small, your board of directors will ask the question. "Do we own anything that is possible to hack by any means ; yes or no?". You are forced to say, "Yes."

    A smaller board question, "Is this new thing impossible to hack ; yes or no?" You are forced to say, "No," if the thing is digital.

    The board is not interested in actual risks. Their questions are for CYA purposes only.

    The point is that you must include the risk of hacking in the cost-benefits to be fully transparent.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Threads for Settingless Relays truly
Alternative to a relay
Switching high voltages with arduino and a relay