Documentation of Changes to FAST Code


1. Changes were made the FAST software code in a controlled manner.
2. The start and end of all changes was bracketed with the following comment lines: !!! Start acr changes and !!! End acr changes.
3. The original FAST code that was not used within the bracketed changes was commented out.
4. Changes to the code would be made available if necessary.

Validation of Changes to FAST Code

1. Results generated with the revised FAST code were compared to results generated using the original FAST code. This was done because some revisions apply to all configurations. Results showed that all output parameters were identical for the two-bladed fixed hub, two-bladed teetering hub, and three-bladed fixed hub configurations.
2. Results generated with the revised FAST code were compared to the expected results for consistency.
  a) One expectation is that the rotating bending moments at the shaft tip about the rotating ya and za axes are expected to be zero with unconstrained teetering . This is consistent with analysis of a two-bladed teetering tub where LSSTipMya remains at zero over the entire analysis. Results confirmed the expectation where both LSSTipMya and LSSTipMza remained at zero over the entire analysis.
  b) A second expectation is that output variables for a teetering, three-bladed turbine would be expected to become very similar to output variables for a fixed-hub, three bladed turbine if the spring constants for the three blades are set exceedingly high. In this case springs were engaged at all times for each of the three blades. Visual comparison of results showed very similar profiles, confirming the expectation.
3. Results generated with the revised FAST code were evaluated for internal consistency. In this case, the initial azimuth position of blade 1 was changed from 0° to 120°. The initial position of blade 1 with this change (120°) is identical to the initial position of blade 2 without this change (120°). Comparison of the two runs showed that all output parameters for blades 1 and 2 were identical. Additionally, output parameters for blades 2 and 3 and blades 3 and 1 are identical.
4. Finally, it was observed during development that even a very small error in the code would rapidly escalate and cause the program to come to an abrupt halt. This is because many variables are monitored during analysis and the program stops running when a variable exceeds allowable limits. The stability of the source code was tested by running a three-blade turbine with teetering enabled for 1000 seconds under extreme wind conditions. The program ran for 1000 seconds without halting.