Analysis of solar magnetic fields using observations as well as theoretical interpretations of the scattering polarisation is commonly designated as a high priority area of the solar research. The interpretation of the observed polarisation raises a serious theoretical challenge to the researchers involved in this field. In fact, realistic interpretations need detailed investigations of the depolarising collisions.

The goal of this paper is to determine new relationships which allow the calculation of any collisional rates of the d-levels of ions by simply determining the value of n* and Ep without the need of determining treating the collisional problem.

In this work, we applied our collisional code to a large number of cases involving complex and simple ions. After that, the results are injected in a genetic programming code in order to infer original relationships which will be of great help to solar applications. We discussed the accuracy of our collisional rates in the cases of complex atoms and atoms with hyperfine structure.

We compared the results obtained through the new relationships with the results obtained directly by running our code of collisions. The percentage of error is about 10% in the average value.