In the paper the authors propose the improvement of three new power quality indexes presented in a previous work [1] for exhaustively detecting voltage sags. In fact, during the indices' experimental assessment, some weak points of the previous presented indices have arisen. After indices' correction, ANOVA analysis have been performed, in order to investigate indices' sensibility with regards to sag depth and duration. At this aim a full experimental analysis has been conducted by applying a multilevel multivariable experiment. Through ANOVA analysis, moreover, a cost function has been realized to correlate the indexes values to the effective cost of lost quality for users who need the absolute absence of sag occurrence. Subsequently the indices' uncertainty has been estimated; a random-fuzzy technique has been preferred, since the ISO Guide shows its limits in presence of DSP-based algorithms. © 2005 IEEE.
Improvement of new synthetic power quality indexes: An original approach to their validation / De Capua, C.; DE FALCO, Stefano; Liccardo, Annalisa; Romeo, E.. - STAMPA. - 2:(2005), pp. 1604247.819-1604247.822. (Intervento presentato al convegno IEEE Instrumentation and Measurement Technology Conference (IMTC), 2005 tenutosi a Ottawa, ON, Canada nel May, 16-19, 2005).
Improvement of new synthetic power quality indexes: An original approach to their validation
DE FALCO, STEFANO;LICCARDO, ANNALISA;
2005
Abstract
In the paper the authors propose the improvement of three new power quality indexes presented in a previous work [1] for exhaustively detecting voltage sags. In fact, during the indices' experimental assessment, some weak points of the previous presented indices have arisen. After indices' correction, ANOVA analysis have been performed, in order to investigate indices' sensibility with regards to sag depth and duration. At this aim a full experimental analysis has been conducted by applying a multilevel multivariable experiment. Through ANOVA analysis, moreover, a cost function has been realized to correlate the indexes values to the effective cost of lost quality for users who need the absolute absence of sag occurrence. Subsequently the indices' uncertainty has been estimated; a random-fuzzy technique has been preferred, since the ISO Guide shows its limits in presence of DSP-based algorithms. © 2005 IEEE.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
