The negative radiation damping may be caused by the diffracting panel elements on the heave plate. It looks like the plate is quite thin, so the panels on the bottom surface will be very close to the panels on the top surface. I guess you have turned on the 'Ignore Modelling Rule Violations' option - when you solved the analysis, there would also have been a warning message '[N] MODELLING RULE #1 VIOLATIONS IGNORED BY GOON OPTION', where almost all of the rule violations would have been due to the panels being very close together.
To investigate this, if you have the time, I would run a mesh sensitivity study. Save a copy of the project; open the copy; reduce the mesh size by a factor of 2 (or more); run the analysis at a single frequency, or a small number of frequencies, where the heave radiation damping is most negative (i.e. around 0.4 rad/s); then compare the results with the original project. Hopefully you will find that the heave damping is less negative with the finer mesh.
I hope this helps!