Load Shedding Scheme 4

[nkumarp]

With a wide range of load shedding devices ranging for less effective under frequency relays with and without df/dt characteristics designed for quick shedding of loads, there still seems to be a lot of options floating around in terms of PLC based and other intelligent devices. I need to conduct loss of generation based transient analysis studies and evaluate subsequently the various options available to finally recommend selection of the most optimum state of the art load shedding scheme for a industrial plant. Would appreciate advice from friends here who have done extensive work on the subject.

 

[gordonl]

Perhaps I misunderstand the question, but if you know how much generation you have and how much load then don't waste anytime. The longer you wait the lower the frequency will drop.

I would add that perhaps you have a small delay to allow the load stabilize. As I've often experienced in these situations, the initial "bump" which causes the df/dt to trip may also cause several loads to trip off line naturally reducing loads, so long enough for this change to be measured would likely be prudent. I would also suggest you should have some form of underfrequency or df/dt as backup to this system as a final line of defense for the most critical loads, ie powerhouse itself.

 

[hsj]

Two types of load shedding are used:
-Frequency relays: By power companies. They have sufficient time to shed the load. If your plant has a generator operting in standalone mode then it is possible to use it. This is not a loss of generation load shedding. It is protection for turbine generators which are sensitive to frequency and also a kind of avoidance of overload.

-Undervoltage relays: Trip everything on loss of generation and start again in steps. Use motorised breakers and contactors in load circuits to switch the load sequentially and insteps. Transient stability study is not required. If your generator is operating in parallel with the power company supply and if there is a disturbance/fault in the power company system then the protection at the tie point should operate to isolate the two systems. If the generator capacity is not adequate for the load then the generator will trip . It is impossible to save it by load shedding. Load shedding cannot be fast enough.

 

[tomatge]

hsj

You're wrong!!

Please read my first post. I'm using a load shedding system to protect a whole island community in the event of failure of the incoming European grid connection. I does (and has) worked sufficiently quickly to allow the generators to keep running.

Provided you can gain access to reliable and fast signalling of a grid failure you can respond with load shedding to keep your own generators running and keep some of your load on supply. The speed at which you do this depends on the characteristics of the generators (speed, inertia etc.) and the power shortfall after the grid supply fails. My advice would be to shed the load as quickly as possible, bearing in mind any reliability considerations. In my case, 100ms is fast enough, but we could improve on this if required, as I said above.

 

[alehman]

One common scheme is to immediately trip blocks of load when separation occurs as suggested by vishalmehta. Blocks can be pre-defined or determined by the system based on actual load and spinning reserve (preferred). Then supplement with 81U and/or measured generator load as a backup to trip additional load blocks if required. I've used this system with diesels for with good results. They are fairly tolerant of momentary overload however.

 

[WattsNew]

To answer this question I agree with nkumarp. Neural Networks is the way to go. ETAP PSMS has a neural network based load shedding system being used in mines, refineries and other industrial processes. Load shedding is done using a PLC but the priority table fed to the PLC is updated continuoulsy based on the system configuration and loading. The key in this system is to not only shed faster but shed an optimal amount. Hope this helps.