Its Ok sir,  sorry for taking up lot of your time.  Thank you for your patient hearing and quick responses.  I shall try some other approach and find some way to it.  Thank you.

Sir ,  Only the inverted 'a' matrix along with 'b' is loaded in to the DSP.  The DSP does only the simple multiplication each time steip. I tried your suggestion, but not much of improvement .  Should I have to give up this method and try some other way, or try to model the circuits as nodes in OM.....will it work.   Kindly guide me.

Thank you

Sir, once again thanking for your immediete responses.  Sorry for troubling you more...

To be clear, from the text file  I get after compiling and optimizing I extract the equations which will be of form ax = b, extract the 'a' matrix find inv(a) and load into the DSP processor b*inv(a) for calculating.  More the equations size of the matrices will be big....I could'nt understand where you are asking me to apply the above solution.

Thank you..

Sir, Thanks for your response.

suppose this is a simple model model circuit of an RC circuit.
import Modelica.Electrical.Analog.Basic.*;
import Modelica.Electrical.Analog.Sources.*;
Resistor R1(R=100);
Capacitor C1(C=0.000001);
SineVoltage V1(V=1,phase=0,freqHz=500);
Ground G;
equation
connect (V1.p, R1.p);
connect (R1.n, C1.p);
connect (C1.n, G.p);
connect (V1.n, G.p);
end circuit;

after using omc +s +d=bltdump *.mo Modelica > *.txt , I get a text file, having 5 variables and 5 equations(optimized) in addition to other things.  The same circuit if I model by MNA will result in only 3 equations. Here only addln  two equations  When I try to model a transmission line of 20 blocks of RLC  number of equations are around 150, where as by MNA it will be around 50.  So the DSP takes lot of time to solve, and hence could'nt meet the real time criteria.  Does OM converts the hybrid DAE to ODE and solves, can we not make it to solve as DAE itself by replacing the reactive elements by current source and conductance.  Please guid me ...

Thank you

Hi,
Basically I am involved in a project of real time circuit simulation in a DSP processor.  We are using OM to model the circuit and load the equations into the DSP processor.
But OM generates huge amout of equations even after optimization.  I need some guidance.  How should I go about?  Can we simplify the equations and reduce the number  further by some means.  Kindly help me...

Thanks

Instead of modeling capacitors as C*der(v) = i  can I model them as current source in parallel with conductance?  can some one guide me on how to write the equations in the model  for that.

Yes sir,  you are right.  Very few equations are left after optimization, but still is it not more than the number of equations formed by nodal analysis or Modified Nodal Analysis.  In spite of optimization I feel some redundant equations are formed. Have I communicated what I wanted to?  I don't know please let me know?

I am not meaning equations involved  while defining the circuit or model.  I mean the equations that are generated by Open Modelica after flattening the model and optimizing it.

For a simple RC circuit with an AC source 3 equations should be suffcicient to solve.  But openModelica genrates around 21 equations.  Even after optimization the number of equation is 5.  How to reduce it further?