International Journal of Scientific Research and Engineering Development

The increasing penetration of distributed generation (DG) in modern distribution networks has improved reliability and efficiency but introduced new challenges to power quality. One of the most critical concerns is the presence of harmonics generated by nonlinear loads and inverter-based DGs. This paper investigates harmonic behavior in hybrid distribution networks under both normal and varying operating conditions. Simulation studies are conducted in MATLAB/Simulink on a photovoltaic–wind integrated distribution system. The study highlights the limitations of traditional harmonic detection methods, particularly in distinguishing between power quality disturbances and unintentional islanding events. A wavelet transform–artificial neural network (WT–ANN) framework is proposed for robust detection and classification. Results indicate that the proposed method provides enhanced accuracy and reliability in harmonic identification compared with conventional techniques. The findings suggest that advanced datadriven models can significantly improve grid stability and monitoring in DG-integrated systems.