Monopiles soil stiffness

[TheMercer]

You say it's a 6 by 6.

So you making one at the soil interface and not several along the span ?.
6 by 6 is an equivalent matrix ?

@SMI-Lab-Inha

[SMI-Lab-Inha]

Good question, and yes on both counts.

It is a single equivalent stiffness at the soil interface (the mudline), not a set of springs distributed along the embedded span. The model is the coupled-spring form from Yu and Amdahl (2023), three springs at the mudline, a lateral spring K_hh, a rotational spring K_rr, and the cross-coupling K_hr between them. The closed-form coefficients (Shadlou and Bhattacharya 2016, via Yu Table 1) already condense the distributed soil reaction over the embedded length into that single mudline stiffness, so the embedment enters through the formulas rather than through node-by-node springs.

Those three terms are then placed into a 6x6 in OpenFAST DOF order [surge, sway, heave, roll, pitch, yaw], with K_hh on the surge and sway diagonals, K_rr on roll and pitch, K_hr on surge-pitch, and -K_hr on sway-roll (opposite sign by the right-hand rule). Heave and yaw are not represented by this surrogate and stay zero. So the 6x6 is an equivalent, condensed matrix at one node, applied at the base of a hub_conn = 3 soft-monopile model via PlatformSupport.mooring_K.

The distributed Winkler bed you are describing (a stiffness spread along the embedded span, distr_k) is a separate, not-yet-implemented path tracked in #118. When that lands you will be able to choose between the condensed 6x6 at the mudline and the distributed springs along the pile.