I have been experimenting with various models for the MFMP Celani replication project and have generated one that shows much promise. The latest model takes the power input variable as its input and then generates both the T_GlassOut waveform as well as the PowerOut one. It does a decent job of calculating these variables as a function of time, following the real world measurements.
One characteristic that seemed odd at first, but makes sense now is that the rising edge of the T_GlassOut waveform has a different dominate time constant than the falling edge. Linear systems will have the same exponential relationship for both. My model demonstrates that this variation is due to the second order dependency of the power output to the T_GlassOut measured temperature. Dave
