How to model a flexibility providing units in the matpower case file

[akhtar3100]

My question is how to model flexible providing units (loads and generation) with different P-Q capability shapes in matpower. For example, if I have a load with active and reactive power values of (50,20) in mpc.bus. What if I want to change these value not rectangularly but in other shape for example have P between 0 and 20 MW, Q between 0 and 5 MVar and P between 20 to 50 Q between 10 to 20 MVAR. How can I do it within the case file and how can i do outside of case file? same for generators.

[rdzman]

I'm going to assume this is in the context of an OPF.

First, let's talk about generators. I assume you already know about the trapezoidal capability curves that are built in to MATPOWER (see Section 6.4.3 in the MATPOWER User's Manual). There is no reason the shape has to be like the one shown in Figure 6-6, with the reactive capability decreasing with increasing output. You simply have the option to define two sloped lines that constrain the capability curve along with the box defined by PMIN-PMAX-QMIN-QMAX.

However, that doesn't sound exactly like what you describe. So, if I understand what you want, I believe you can accomplish it by modeling your generator as two separate generators, with different reactive ranges. You will want to use a slightly lower active power cost on the first generator to ensure that it get's dispatched first. Although, now that I think about it, if reactive power needs are great it could still choose the 2nd unit before the first if the active power cost difference is not great enough.

For the loads, you can simply treat them as negative generators, i.e. move them from the bus matrix to the gen matrix. The load2disp() function can help you with this. However, there is one important caveat here. MATPOWER treats any generator with PMIN < 0 and PMAX == 0 as a special case dispatchable load with a constant power factor constraint added (see Section 6.4.2 in the MATPOWER User's Manual). In your case, you do not want this constant power factor constraint. You can eliminate it by setting PMAX on these generators to a negative number (e.g. -eps).

[akhtar3100]

Thanks, I think for my case the 3rd case of dispatchable laods is relevant. And I think load2disp() would be helpful. However, i have two question, for constant power factor dispatchable load how the powerfactor value is determined. how to provide constant powerfactor in that case. Second when we don't want constant power factor constraint for example and want to have a circular P-Q region for dispatchable loads how to do that.

Regards, Akhtar

[akhtar3100]

Furthermore, Static generators are model as negative loads right? what if I want to have static generators with different Q capabilities constant power factor, constant Q mode, how to model that?

[rdzman]

... for constant power factor dispatchable load how the powerfactor value is determined. how to provide constant powerfactor in that case.

It is determined by the ratio of PMIN and either QMIN or QMAX, whichever is non-zero. (At least one of them must be zero).

Second when we don't want constant power factor constraint for example and want to have a circular P-Q region for dispatchable loads how to do that.

MATPOWER does not implement a circular PQ region. For that you would need to implement your own non-linear constraint See Section 6.3.2 in the MATPOWER User's Manual.

[rdzman]

Furthermore, Static generators are model as negative loads right?

They can be modeled as negative loads (in the PD and QD fields of the bus matrix), or they can be modeled as generators with PMIN = PMAX = PG and QMIN = QMAX = QG.

what if I want to have static generators with different Q capabilities constant power factor, constant Q mode, how to model that?

In the context of an OPF, you simply define the "static" generator's active and reactive power limits based on the capabilities you desire. For a "static generator" I assume the active power output is fixed. If power factor is fixed, then so is the reactive output (QMIN = QMAX).

[akhtar3100]

It is determined by the ratio of PMIN and either QMIN or QMAX, whichever is non-zero. (At least one of them must be zero). bus Pg Qg Qmax Qmin Vg mBase status Pmax Pmin Pc1 Pc2 Qc1min Qc1max Qc2min Qc2max ramp_agc ramp_10 ramp_30 ramp_q apf 5 -0.02 -0.005 0 -0.005 1 100 1 0 -0.02 0 0 0 0 0 0 Inf Inf Inf Inf 0 So far this case for a dispatchable load Pg=-0.02, Qg=-0.005, Qmax=0, Qmin=-0.005, Pmin=-0.02 would powerfactor be pf=Pmin/Qmin=-0.02/-0.005=4, but that does not make sense and P/Q is normally inverse tanget. I think the power factor should be calculated as pf=P/(P^2+Q^2), wht do you think? MATPOWER does not implement a circular PQ region. For that you would need to implement your own non-linear constraint See Section 6.3.2 in the MATPOWER User's Manual. ok

[akhtar3100]

Furthermore, Static generators are model as negative loads right?

They can be modeled as negative loads (in the PD and QD fields of the bus matrix), or they can be modeled as generators with PMIN = PMAX = PG and QMIN = QMAX = QG.

what if I want to have static generators with different Q capabilities constant power factor, constant Q mode, how to model that?

In the context of an OPF, you simply define the "static" generator's active and reactive power limits based on the capabilities you desire. For a "static generator" I assume the active power output is fixed. If power factor is fixed, then so is the reactive output (QMIN = QMAX).

What if I want to set powerfactor limits for the static generator for powerflow? What I want to do is that I want to have P-Q boundaries defined for static generators and then randomly choose (p,q) points within this boundary for powerflows.

[rdzman]

I think the power factor should be calculated as pf=P/(P^2+Q^2), wht do you think?**

Sorry if I was unclear. I did not mean that the power factor is equal to PMIN/QMIN, simply that the values of PMIN and QMIN, particularly their ratio and the implied angle, define the power factor.

[rdzman]

What if I want to set powerfactor limits for the static generator for powerflow? What I want to do is that I want to have P-Q boundaries defined for static generators and then randomly choose (p,q) points within this boundary for powerflows.

In this case P and Q are fixed within the context of the power flow computation, which does not need to know anything about the P-Q boundaries. You simply use the process you describe to select the input values for the power flow. This process does need to know about the P-Q boundaries. Right?

[akhtar3100]

yeah. Right. I will create a process which defines these boundaries and then run power flow. I thought maybe there was some function to do it.

Thanks