SKM Load Flow with PV Generators

[ThePunisher]

In SKM manual, when running multiple on-site generators that are connected away from the utility:

"An on-site generator that operates in parallel with the utility source may be defined as either a PV- or PQ-type generator..The swing bus generator or utility source must supply any deficiency in generation, and all losses. If there is excess generation in the system, the source bus acts to absorb the excess generation.Usually, a system requires only a single swing bus. Multiple swing buses may be used to model large systems having multiple connections to a utility. Therefore, the exact utility voltages and angles may be specified. Additionally, multiple independent systems require that you specify a swing bus for each system. Only one swing bus may be specified at a single bus". Co-generation modeled as PV-type generators which are not located near the utility or swing bus generator will serve to regulate the voltage at the generator bus. PTW attempts to hold the bus voltage at the target level within the reactive power range. It may not be possible to obtain the target voltage. PTW will maintain the kW output of the generator and allow the voltage to float to the reactive power limit".

I am a bit confused with the part that says: "PTW attempts to hold the bus voltage at the target level within the reactive power range. It may not be possible to obtain the target voltage. PTW will maintain the kW output of the generator and allow the voltage to float to the reactive power limit"

What does this mean?

I have a little bit of dilemna on the load flow because, if the in-plant generation exceeds the load by a couple og MWs, the power flow is back-feeding to the transformers. In reality, we intend to maximize and adjust the on-site generation in such a manner that i will limit my electric utility costs and during an operation where the utility and generation are on-line, I optimize my generators to limit my utility consumption. Is there a way I can do that in a load flow study where the calculation is static and not dynamic.

 

[dpc]

Just set the generator kW output to match the load. For a generator modeled as a PV, you should be able to set any generator output you want. After you run your power flow, you will know what the plant load is - for this particular load condition. Just set the generator kW to match.

Also, you are not optimizing your generators - they don't care what your plant load is. You are optimizing your fuel consumption. If your plant load is highly variable, your prime movers may not like swinging load constantly. The generators really won't be bother much.

Also, what are you trying to figure out by running a power flow analysis? That will determine how you model it.

 

[ThePunisher]

Thanks for the reminder DPC. I am actually running load flow to check loading-wise if the plant generation and utility are well within limits during normal and contingency situations. I am using load flow since it will also consider system losses.

I did specified an operating generator MW output to be a little lesser than the MW loading (with the utility on-line and available). I still have to put a MVAR range on the generator component editor so the program will have an MVAR range to play with to maintain targeted voltage.

My next question would be, assuming the generator is having a rated power factor of 0.80 and I already know the targeted generator MVAR output (to be little lesser than the load MVAR requirement). What would be the "other" estimated MVAR so I cna specify the range? Can I say I get the MVAR based on an 80% P.F. up to the target MVAR?

 

[dpc]

The range you specify doesn't really matter in the end. You'll have fewer issue with convergence if you give it a wide kVAR range to iterate through. If the solution ends up outside of the generator's actual reactive power limits, you know you have a problem - you'll have to tweak the voltage setpoint a little. A small change in voltage setpoint can cause a huge kVAR swing. I can't remember if PTW allows you to specify a generator power factor instead of a voltage setpoint. Also, be aware of the actual tap settings for any transformers in your system as well as the impact of any LTC or step-type voltage regulators. Any cap banks also need to be modeled.

The MVAR in the solution is what it is - it doesn't depend on the range you specify, but rather on the reactive power needs in the loads you have modeled.

 

[ThePunisher]

Thanks for your assistance and patience.