Symmetric coin cell cycling is an important tool for the analysis of battery materials, enabling the study of electrode/electrolyte systems under realistic operating conditions. Understanding the behavior of metal anodes in batteries and accurately predicting their performance is a challenge due to the methodological gap between theoretical models and experimental observations. To address this challenge, a PDE model describing the time evolution of voltage profiles under the Galvanostatic Discharge-Charge (GDC) protocol in Li/Li symmetric cells has been developed [1, 2]. In this talk, based on [3], we propose a hybrid architecture combining Convolutional Neural Networks and Long Short-Term Memory layers (CNN-LSTM) to estimate key physico-chemical parameters in the PDE system governing the GDC cycling process. Our results show the ability of the network to capture complex temporal characteristics of voltage profiles – such as peaks and valleys, saddle points, and changes in concavity [1] – whose features are often missed by conventional approaches such as Least Squares (LS) fitting. Moreover, our deep learning model can generalize also to experimental discharge charge time series data. These results highlight the capability of our CNN-LSTM neural architectures in bridging the gap between physical modeling and experimental measurement for time-dependent battery systems.