Use of PROII for accelerated convergence
Thread Content
The last edit to this post was made by wei_huacn on 2019-3-8 at 15:28. In the proii software, for the options related to problem recurrence, convergence, and acceleration, there are default values for parameters such as temperature and pressure. Sometimes, when convergence is difficult, I increase these values; however, I’m not entirely sure of the principle behind this. Is it true that the smaller these values are, the closer they are to the actual values, making convergence more difficult? Also, what is the difference between the acceleration options Wegstein and Broyden?
Do you mean that the default value refers to the allowable error? A larger allowable error will certainly make convergence easier. Different convergence methods have different convergence speeds.
Wegstein and Broyden are different iterative calculation methods; just as there is a difference between binary search and Lagrange interpolation, their convergence rates vary in different situations
One is the convergence error, and the other is the acceleration option. What is the difference between Wegstein and Broyden? They are options available under Acceleration Options. What is the difference between these two convergence methods and the No acceleration option?
I see, but in what situations are these two iterative calculation methods applicable?
The help file mentions that if one method doesn’t work during actual use, try another one; give it more attempts
In Sun Lanyi’s second edition of \"Chemical Process Simulation Practice\", pages 435–438, various convergence methods are described in detail; it’s worth taking a look.