Erratum for “Krylov Subspace Accelerated Newton Algorithm: Application to Dynamic Progressive Collapse Simulation of Frames” by Michael H. Scott and Gregory L. Fenves

Fig. 2 of the original article contained two errors. A revised Fig. 2 is included herein and corrects those errors as follows:

First, Line 10 of the original pseudocode in Fig. 2 sets the stiffness matrix, $\mathbf{K}$, equal to the Jacobian matrix. This terminology implies a sign inconsistency with the system of equations shown in Eq. (7) of the original article. The correct terminology is stiffness, as shown in the corrected figure.

Second, the authors omitted a line of code from the original pseudocode. The algorithm requires Line 26 in the corrected figure in order to compute the $\mathbf{w}$ vector when the subspace dimension, $m$, is greater than zero. Computation of the $\mathbf{w}$ vector is shown on Line 23 in the corrected Fig. 2 and described in Eq. (13) of the original article.