Neural Networks Based Dynamic Load Modeling for Power System Reliability Assessment

The reliability of a power system is considered as critical requirement in planning and operating the due to increasing demand for more reliable service with lower frequency duration interruption. Hence, also major challenge development future systems they become advanced complex, making accuracy assessment dependent on several factors such supply load modeling. Recent studies systems’ stability have focused modeling, where loads are either assumed be static or dynamic, by introducing significant constraints. However, emergence new types necessitates models that can incorporate them accuracy, this would facilitate their effective use flow simulation studies, well analyses. In study, dynamic modeled using feed-forward neural network test bed developed MATLAB/Simulink generate data used during training validating model. Subsequently, Electrical Transient Analyzer Program (ETAP) software verify effect modeling platform. Bus 2 Roy Billinton Test System (RBTS) employed case study investigate sensitivity indices, Average Interruption Duration Index (SAIDI) Frequency (SAIFI), technique mixed (dynamics statics).