Radius of Convergence
- Type
Floating point number
- Range
[1e-7,1]
- Default
0.01
This option determines when the penalty parameter \(\rho\) will be reduced (if initialized to a positive value). Both the nonlinear constraint violation (see option Row Tolerance) and the relative change in consecutive Lagrange multipler estimates must be less than the value of this option at the start of a major iteration before \(\rho\) is reduced or set to zero. Once \(\rho\) is zero, the sequence of major iterations should converge quadratically to an optimum.
Learn more about