The usage of Lithium-ion (Li-ion) batteries has gained widespread popularity across various industries, from powering portable electronic devices to propelling electric vehicles and supporting energy storage systems. A central challenge in Li-ion battery reliability lies in accurately predicting their Remaining Useful Life (RUL), which is a critical measure for proactive maintenance and predictive analytics. This study presents a novel approach that harnesses the power of multiple denoising modules, each trained to address specific types of noise commonly encountered in battery data. Specifically, a denoising auto-encoder and a wavelet denoiser are used to generate encoded/decomposed representations, which are subsequently processed through dedicated self-attention transformer encoders. After extensive experimentation on NASA and CALCE data, a broad spectrum of health indicator values are estimated under a set of diverse noise patterns. The reported error metrics on these data are on par with or better than the state-of-the-art reported in recent literature.

Sonification uses sounds to glean insights about information and activities in a person’s life. There are two types of emotions based on sounds: perceived emotions and induced emotions. This paper focuses on classifying perceived emotions based on two dimensions – arousal and valence, using several deep-learning models. Four feature selection techniques, Forward Feature Selection, Recursive Feature Elimination, Random Forest, and Principal Component Analysis, are performed; class imbalance in the dataset is demonstrated and handled using under-sampling, and over-sampling techniques, and the results are compared. This paper shows the need for balanced data to train classifiers and the advantages of running classifiers on the balanced dataset that is generated using sampling techniques. The eXtreme Gradient Boosting (XgB) classifier trained and tested on the over-sampled balanced dataset using all the features generates a test F1 score of 81.5 and is the best model that can be selected from all the classifiers.