The significance of dynamic simulation models for estimating the behaviour of combined cycle power plants is widely acknowledged. In part B of this study, the dynamic behaviour of a sub-critical three-pressure- stage heat recovery steam generator is numerically investigated, employing the process simulation tools Aspen Plus Dynamics and Apros. A wide set of measurement data from an existing power plant is applied to validate the generated dynamic models during warm start-up procedure. The numerical results follow qualitatively the transients during initial start-up phase and show good agreement during the post start-up phase. Furthermore, similar qualitative behaviour between both process simulation codes used is observed. The obtained results suggest that the dynamic simulation models are capable to reliably predict the system response to failure malfunctions and to modifications in design and control structures. The provided measurements are of high relevance for other researchers due to the fact that such detailed experimental data is proprietary and rarely published in scientific literature. (C) 2015 Elsevier Ltd. All rights reserved.