1. You should add it to Sbus in F(x0) and F(x). 2. You can either add it inside dSbus_dV.m, or you can modify the return values directly in newtonpf.m.
3. First, regarding notation, just to be clear, what you wrote as Vi and Vj are simply Vm(i) and Vm(j). Likewise θi and θj are simply Va(i) and Va(j). And dSbus_dV.m just gives a compact form for computing the partial derivatives using complex matrix notation. But you can write the partial derivatives of each of your functions (Pi, Qi, Pj, Qj) with respect to, and in terms of Vm(i), Vm(j), Va(i) and Va(j). Then you simply evaluate those partial derivatives using the current values of Vm and Va and modify the appropriate elements in dSbus_dVm and dSbus_dVa. For example, for Pi, you will have 4 partial derivatives which will be added to (or actually subtracted from, I suppose, since Sbus has a negative sign in the mismatch equation) the real part of the corresponding two elements in row i of dSbus_dVm and two elements in row i of dSbus_dVa. For Qi you’ll modify the imaginary parts of the same entries. For Pj and Qj, the elements will be those in row j. Hope this helps, -- Ray Zimmerman Senior Research Associate B30 Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Nov 26, 2013, at 10:05 AM, hussein Adel <[email protected]> wrote: > I can summarize the issue in 3 points: > Adding this variable P and Q means I have to add them to the jacobian as well > as the mismatch equations. > > 1- If I will add them to the mismatch equations, can I add them to Sbus in > F(x) only? or Sbus in F(x0) and Sbus in F(x)? > > 2- If I have the power equation for example as Pj = constant*Vi*V j *Sin(θi > −θ j +constant) and similar equations for Pi,Qi,Qj. > newtonpf.m calls this code [dSbus_dVm, dSbus_dVa] = dSbus_dV(Ybus, V); > I think to add those P and Q, I have to add their partial derivatives to that > dSbus_dV file then, is that true? > > 3- I don't understand at all how I can convert the value or Pi for example to > a partile derivative W.R.T. Vm or Va, even though I understand the partial > derivative equation that already exists in the dSbus_dV file. > From: [email protected] > Subject: Re: Editing the Jacobian Matrix > Date: Tue, 26 Nov 2013 09:38:36 -0500 > To: [email protected] > > Ok … so, is there a specific part of newtonpf.m that you don’t follow? It > seems to me that you simply need to find and modify the corresponding element > in F and row in J. Is the difficulty in finding the appropriate elements to > modify? or knowing how to modify them? > > -- > Ray Zimmerman > Senior Research Associate > B30 Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > On Nov 25, 2013, at 9:11 PM, hussein Adel <[email protected]> wrote: > > I am not sure if replying this mail is the right way to put my reply on the > archive, but I am sorry if it is wrong. What I meant by editing the jacobian > matrix is: > I want to add a device whose P and Q depend on the voltage magnitude and > angle. This is why I want to insert it in the jacobian and mismatch > equations. But this is to be done at only one bus. This is what I do not know > how to do even though I tried many times with newtonpf.m , runpf.m , and > dSbus_dV.m > > I hope it is clear, and I hope you can answer me. > Regards. > > From: [email protected] > Subject: Re: Editing the Jacobian Matrix > Date: Mon, 25 Nov 2013 16:34:49 -0500 > To: [email protected] > > I’m not sure I understand what you want to do. If you mean that you simply > want to modify the fixed net real and reactive power injections at a given > bus, you can do this simply by changing the values in the PD and QD columns > in the bus matrix. If you mean that you want to model some new devices whose > P and Q depend on other variables (e.g. bus voltage), then you will need to > get into the code, and anything is possible depending how deep you want to > get. Have a look at newtonpf.m to get started. > > -- > Ray Zimmerman > Senior Research Associate > B30 Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > On Nov 22, 2013, at 5:29 PM, hussein Adel <[email protected]> wrote: > > I want to add components on the 14 bus system that exists in MATPOWER. The > problem is that inquires me to add P&Q at specific two buses. The addition of > those P&Q has to be done at the jacobian matrix and the mismatch equations > directly. It is so unclear for me how to add components at specific buses on > the jacobian matrix or the mismatch equations in matpower. > It seems to me like it is done as a whole and I cant make a change at a > specific bus alone in the matrix. > Any help is appreciated.
