The main goal of my bachelor thesis was determining those diagnostic parameters which are capable of estimating the ageing of EPR and XLPO insulated signal cables. For that I measured three different parameters of thermally aged cable samples: dissipation factor in different frequency ranges, initial slope of return voltages with different time constants and the initial slope of discharge voltage.
After running the date through regression analysis, I found that in the case of EPR insulation it is best if the dissipation factor is measured, especially in the 20 Hz ~ 2 kHz frequency range. After that an estimation model can be erected. In my case this model had a relative error of less than 12%. In the case of XLPO insulation it is better however to measure the initial slope of return voltage in the 30s ~ 300s and 750s ~ 1500s time constant ranges. This model had a slightly worse accuracy with a relative error of 15 ~ 20%.
A strong, unequivocal connection between the ageing of the cables and the initial slope of discharge voltage could not be stated in either case of insulations.
My thesis also contains relevant literature about insulations (ageing and deterioration of insulations, measuring methods of the parameters mentioned above) and the statistical method used to determine the appropriate diagnostic parameters, regression analysis.