AI-DRIVEN RELIABILITY ENHANCEMENT OF SMART POWER GRIDS USING RANDOM FOREST-BASED TRANSMISSION LINE FAULT PREDICTION

Transmission line failures pose significant risks to the reliability and safety of electrical power systems, leading to costly outages and system downtimes. Early and accurate detection of potential transmission line failures is essential for ensuring uninterrupted power delivery and efficient maintenance planning. In this project, we present a robust classification framework for transmission line failure prediction using ensemble machine learning techniques, focusing on improving predictive accuracy and model robustness. The study begins with comprehensive Exploratory Data Analysis (EDA) to understand the underlying patterns and relationships in the dataset. Through statistical summaries, correlation analysis, and visualization techniques, important features influencing transmission line failures are identified. This analysis plays a crucial role in refining the dataset and enhancing model performance. As part of the machine learning modeling process, a Decision Tree Classifier is initially employed as the existing system. While it offers interpretability and simplicity, the model tends to overfit and may lack generalization capability when dealing with complex or noisy data. To overcome these limitations, we propose the use of a Random Forest Classifier as the proposed system, leveraging the power of ensemble learning. By aggregating the predictions of multiple decision trees, the Random Forest model significantly improves accuracy, reduces overfitting, and provides better handling of feature interactions. The results demonstrate that the Random Forest Classifier outperforms the Decision Tree Classifier across all major metrics, establishing its effectiveness in classifying transmission line failures with higher reliability. The project highlights the advantages of ensemble methods in critical infrastructure applications and showcases how data-driven techniques can be harnessed to develop intelligent fault detection systems in the power sector.