Classification of voltage sag disturbance sources using fuzzy comprehensive evaluation method

The classification and recognition of voltage sag disturbance sources is the foundation of mitigating voltage sags. This study analysed characteristics of various kinds of voltage sags, including sags caused by short-circuit faults, energising of transformers and starting of large induction motors. According to the features of different sources, an evaluation index set was formed by three-phase voltage unbalance factor, duration ratio and increments of second harmonic. After that, the weight of each element was calculated by analytic hierarchy process. A novel kind of estimation criterion for fuzzy membership function parameters was proposed, inspired from 95% probability value, based on the method of undetermined coefficient to establish the functions. Consequently, according to it, fuzzy membership functions were created for each index. Further, comprehensive index membership degrees were calculated to judge the disturbance source according to the maximum of the degrees. The presented estimation criterion can efficiently reduce the influence of sample errors and thus improve the accuracy of the whole process, which was verified with the use of data obtained from both simulations and measurements.