Abstarct: Power quality is a critical factor influencing the monitoring of electrical distribution grid performance, and disturbances in power quality can significantly affect the operation of electrical equipment and the satisfaction of consumers. This thesis investigates methods for detecting and classifying power quality disturbances in distribution networks through the application of active deep learning techniques. The primary objective of this research is to develop effective methodologies for identifying and categorizing various types of power quality disturbances, including voltage fluctuations, harmonics, and imbalances in current and voltage. For this study, an industrial active distribution grid associated with the power infrastructure of Alay Mehestan Petrochemical Company has been selected. This company, situated within the Petro Refinery Complex of Kangan, boasts an annual polypropylene production capacity of 450,000 tons. Additionally, the research considers a solar power plant unit dedicated to electricity generation. The network is modeled utilizing ETAP software, encompassing all consumers, and is subsequently subjected to load flow and harmonic studies. Artificial Intelligence algorithms and signal analysis techniques are employed to analyze the data obtained from these studies to identify disturbance patterns. The findings of this research indicate that the proposed models demonstrate a high level of accuracy and rapidity in detecting various types of disturbances, which significantly enhances the overall power quality within distribution networks. Moreover, the results emphasize that timely detection and classification of power quality disturbances can support engineers and network designers in implementing suitable strategies to manage and enhance power quality. This research could serve as a foundation for the advancement of intelligent systems aimed at power quality management and the optimization of performance within active distribution networks in various large-scale industries, including petrochemical, refinery, mining, steel, copper, and others.