<p>In present times, where the entire world is taking a paradigm shift from conventional fuels to Li-ion batteries, there also exist several challenges. The major challenge being approximating the degradation of battery life. Due to several factors, the degradation is subjected to high non-linearity, making it difficult for decision-makers to take appropriate measures for battery life cycle management, including approximating the battery’s remaining useful life (RUL). Thus, based on the aforementioned lacunas, this study proposes an inventive hybrid methodology that can approximate the RUL of a battery using its operating parameters. The approach combines two well-established models to form a hybrid one-dimensional Convolutional Long Short Term Memory (1DConvo-LSTM) model that can perfectly mimic the degradation pattern of the battery, thereby accurately predicting its RUL. The study proposes an experimental investigation into identifying the best suitable modeling and hyperparameters, the combination of which can efficiently improvise the prediction results. The research includes three&#xa0;experimental cases. The first case identifies the best optimizer among set of optimizers. The second case applies a feature selection technique with the best optimizer identified from the first case. The third case incorporates dimensional reduction via PCA. Additionally, the results of the proposed model are also compared with a few of the benchmark algorithms. The findings suggest that 1DConvo-LSTM model with AdamW optimizer, 4 features, namely, CI, M_xVD (V), M_nVC (V), and CT, results in best performance measures with RMSE, MAE, MSE, R-squared value, and convergence time of 0.0040, 0.0017, 0.0001, 0.999 and 360s respectively.</p>