Hello all,
I have a code (in python) which is run a dynamic model in PSSE. For every simulation, I run the code and get the P and Q of the generator ( I set the final time of simulation, as 20 seconds). Every simulations takes 3-5 seconds to complete. Does anybody now what are the factor that impact this time? and does any body can help me to how to reduce this time? Thank you. best regards
Standard simulations parameter values for integration step size is 1/4 of a cycle (60 Hz), acceleration factor set to 1. Verify that during case conversion, TSYL activity converged in 1 or 2 iterations. Select minimum but necessary set of channels. Send output of process to a log file (or to null device) instead of the default screen. Try running the simulation outside the PSSe GUI, using python and PSSe API. These settings will give you a baseline simulation time. Deviations from these settings may render longer simulation times.
For a dynamic run, 3-5 sec total simulation time is fast. If you have to do multiple runs (>100) consider parallel processing. Search this forum for more related posting.
Thank you so much. Your comments is useful for me. But I don't understand what is the acceleration factor in the code. pssspy.run(option, tpause, nprt, nplt, ctrplt) here is the setting for me pssspy.run(0, 2, nprt??, nplt??, ctrplt??) I dont know how to set the three parameters in this line.
From the API.pdf: acceleration factor is set to 1.0 as, psspy.dynamics_solution_param_2([99,_i,_i,_i,_i,_i,_i,_i], [ 1.0,_f, 0.001, 0.004,_f,_f,_f,_f]) The run command writes data to disk by skipping 3, to screen/log by skipping 99, plot 0 points: psspy.run(0, 2, 99, 3, 0) #nprt?, nplt?, ctrplt?
Asked: 2020-05-09 17:41:06 -0500
Seen: 607 times
Last updated: May 10 '20
whit loves you. Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
