The block subspace iteration method is recommended for calculation of large structures with an excessive amount of modes to be calculated.
The subspace iteration methods solve the equation system of eigenvalue problems Kφ - λBφ = 0.
This method is usually quicker than the Lanczos method.
The tolerance (required convergence) of the eigenproblem solution is determined by the formula:
The tolerance parameter (tol) is defined in the Modal Analysis Parameters dialog. The accuracy of calculations can be increased by reducing the tolerance tol value. This results in the growth of the number of iterations.
Calculations
The calculations dialog for the Subspace Iteration method shows the following phases:
- Number of the current iteration / maximum number of iterations.
- Required accuracy (tolerance).
- Accuracy of the current iteration.
- Number of required modes.
See also: