Bismuth-Based Double Perovskites for Solar-Powered EVs using ResNet and Long Short-Term Memory

As people rely more on cars for transportation, greenhouse gas emissions and ozone layer depletion rise. For shorter rides, solar-powered e-bikes are affordable and environmentally friendly. This proposed approached a ResLSTM hybrid model with LSTM. The methodology involves preprocessing, feature extraction, and model training. Remove duplicates and missing values before k-means grouping and training. Consistency requires scaling, normalizing, and standardizing features. With an average accuracy of 93.45%, the ResLSTM model surpassed current approaches. The recommended method effectively estimates solar energy and vehicle energy consumption to maximize energy use for solar-powered e-bikes. The results suggest that ResLSTM can provide efficient and eco-friendly transportation options. Resolving data anomalies and ensuring accurate projections helps solar-powered electric bikes become more popular as a sustainable transportation choice.