During Covid-19 lockdown I used Dr, Ware’s Simple Cyanotype frequently at DR 2.70 which is a bit difficult with the highlights not clouding but used Piezography Professional Editions Toolkit to linearize it quite successfully in just two iterations using the beginning Cyanotype.quad
I can’t know all that you are trying to accomplish with your spreadsheet you sent - but I wonder if you have not had success with using our tools and therefore trying to attain something you can’t with the toolset…
We usually direct that standard exposure time is attained first by simple exposure step test through clear pictorico OHP to find first optical dMax. Then use base cyanotype.quad to begin calibration with target which we prefer is 21x6 for hand applied sensitizers. Measure create new curve to print same target. Measure create 2nd curve and based on how linear the results were on the 2nd iteration make the decision to use as final or do one more validation test with new curve/target to create final curve.
Cyanotype is also affected by humidity so this should all be done in controlled environment. Takes just a few hours from beginning to end - drying is the slowest part…
Are you doing something different from this? For example your base exposure is it dependent on factors other than exposure through film - perhaps some sort of example following another type of workflow and then you try bring PiezoDN in on the equation? Have you tried pure PiezoDN workflow?