Lithium-ion battery cycling and aging measurements collected under tropical environmental conditions in Cameroon
Description
This dataset was generated to characterize lithium-ion battery degradation in the context of electric vehicle operation under tropical environmental conditions. The working hypothesis underlying the data generation process is that sustained exposure to elevated ambient temperature and humidity, combined with repeated charge–discharge cycling typical of electric mobility usage profiles, accelerates electrochemical aging mechanisms and modifies the dynamic electrical and thermal response of the battery system. The batteries were operated following controlled cycling protocols representative of electric vehicle usage, including repeated charge–discharge sequences at defined current rates. Data acquisition was performed using calibrated battery testing systems and integrated monitoring sensors. Continuous time-series measurements were recorded for voltage, current, and surface temperature. Cycle-resolved variables include discharge capacity, internal resistance estimation, cycle index, depth of discharge, and health-related indicators derived from capacity evolution. Environmental parameters reflecting tropical operating conditions were monitored during the acquisition period to account for thermal stress exposure. The dataset includes both raw time-series signals and processed cycle-level summaries to enable multi-scale analysis. The data exhibit progressive capacity fade, increasing internal resistance trends, and evolving voltage response profiles across cycles. Thermal behavior during high-current operation phases shows measurable variations consistent with aging progression. These patterns allow quantification of degradation trajectories under electric vehicle operational constraints. The dataset is structured to support advanced modeling tasks including time-series forecasting, degradation curve fitting, multivariate regression, dimensionality reduction, regularization techniques addressing correlated predictors, and deep learning architectures for health estimation. It can also be used for benchmarking predictive maintenance algorithms, probabilistic graphical modeling, reliability analysis, and algorithm validation under high-temperature electric vehicle operating conditions. The data are organized in structured spreadsheet format with clearly defined units and variable descriptions to facilitate reuse in electrochemical modeling, data-driven health estimation, and energy storage system reliability research.
Files
Institutions
- University of DoualaLittoral, Douala