Results for aqueous glucose system
The table below compares the actual parameter values with estimates obtained using three starting guesses. The guessed values were used for all five of the parameters.
The results show that the estimation routine is sensitive to the guess values. In my experience thus far, one of the order of magnitude guesses usually results in an acceptable answer. In the cases above, two of the outcomes (guess = 1, 0.1) were obviously incorrect. In some cases, it is not obvious from the numerical values which outcome is most correct. In those cases, the plot showing the goodness of fit to the hydrogen flow rate can be used to select the best answer. The figure below shows the fit for the guess = 10 outcome.
I am going to postpone the study of error effects for later because I want to select a system with an organic solvent and a "known" kinetic model. Stay tuned using the RSS feed!