PROJECT TITLE :
Improved DGA method based on rules extracted from high-dimension input space
The diagnosis of incipient faults in power system parts like transformers is sometimes based on the concentrations of dissolved gases existent within the insulation oil. There are consolidated DGA-based (dissolved gas analysis) methods within the literature, like the Duval triangle. But, they present some limitations like the existence of non-call areas and erroneous results. Proposed may be a straightforward methodology to improve the analysis of incipient faults primarily based on rules extracted from a high-dimension space (twenty one attributes), fashioned by the gases concentrations and some of their interrelations. From such input space, the C4.5 technique (call tree) is employed to extract a set of interpretable rules. Databases known within the DGA technical literature like IEC TC 10 are adopted to analyse the proposed approach. Compared with a customary technique, considering all data test folders in the performed 10-folder cross-validation statistical analysis, the extracted rules show larger accuracy with a slip-up in the diagnosis of incipient faults of six.twenty five%, against eighteen.seventy five% for the Triangle method within the worst case.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here