Analysis of drawbacks and constraints of classification algorithms for three-phase voltage dips

Abstract

Voltage events are one of the most common and harmful disturbances of power electric systems. Voltage dips, swells and interruptions are included under this heading. Given the economic cost that these disturbances represent for electrical power transmission and distribution companies and the industry, it becomes imperative to detect and classify them properly. Several classification criteria and algorithms have been proposed in the literature as analysis tools to differentiate voltage events by their characteristics and, if possible, to determine their causes and consequences. Even though some of these approaches make a correct classification of the voltage events, there are certain operation conditions that are common in real electrical grids, in which the classification criteria, and their corresponding algorithms, make a wrong classification. These particular conditions, together with the lack of a fair comparison in a common scenario, have not been addressed in the specific field literature. This work explores in detail all these aspects by evaluating the symmetrical components criterion and ABC classification criterion, and rigorously analyzes three specific algorithms: the Symmetrical Components Algorithm, the Six Phases Algorithm and the Space Vector Algorithm. Drawbacks arise from both classification criteria and algorithms. The causes of the classification errors are described and discussed in detail in order to better understand the problem, and evidence the constraints of these classification methods.