AI-Powered Energy Management Systems for Hybrid Electric Vehicles using Advanced Machine Learning Models

Hazardous substances released into the atmosphere by humans are producing a global environmental crisis. Fossil fuel combustion drives industrial and transportation developments. Despite being vital to global development, fossil fuels are scarce and in high demand, raising costs. These environmental concerns require innovative solutions and cutting-edge technology. This study uses online data collection, HDP preprocessing, and Enhanced Marine Predators Algorithm feature extraction. A new machine learning technique based on the XG-LGBM (Extreme Gradient Boosting and Light Gradient Boosting Machine) identified and categorised erroneous battery data to increase BMS safety and dependability. Python simulations used statistical analysis, deep learning, ML, and experimental validation. The XG-LGBM-based method identified and classified inaccurate battery data better than existing methods. Model accuracy and efficiency enhanced by EMPA feature extraction. Simulations demonstrated the technique enhanced BMS reliability and safety. This study highlights how machine learning may solve environmental and technological issues. XG-LGBM makes energy management systems safer and more sustainable with its powerful BMS reliability solution. This method fixes battery data inaccuracies to reduce environmental impact and increase resource efficiency.