A Fusion Mechanism-Coordinated Dynamic Modeling Approach for Smartphone Battery Depletion Prediction

Accurate time-to-empty (TTE) estimation for smartphones remains challenging because battery electrochemical dynamics interact with highly variable device-level power demands. This study develops the Fusion Mechanism-Coordinated Dynamic Model (FM-CDM), a physics-informed theoretical framework that couples a second-order Thevenin equivalent-circuit model with component-level power consumption, electro-thermal feedback, battery state of health, and stochastic workload generation. TTE is formulated as the first time at which the terminal voltage reaches a 3.2 V cutoff, thereby distinguishing the shutdown condition from SOC = 0%. The algebraic coupling among device power, discharge current, and terminal voltage is resolved using the physically admissible solution of the constant-power load equation. Separate battery and processor thermal states are introduced to represent temperature-dependent internal resistance, battery heat generation, heat dissipation, and processor thermal throttling. Four representative workload classes—Standby, Light, Medium, and Heavy—are considered, with 500 Monte Carlo realizations used for each class to propagate workload and parameter uncertainty. Global sensitivity is evaluated using the Morris elementary-effects method. The illustrative numerical analysis shows that increasing component activity shortens model-estimated runtime and that voltage-triggered shutdown can occur at a nonzero residual SOC, particularly when temperature-dependent resistance and load-induced voltage drop become significant. The framework provides a transparent and reproducible basis for investigating smartphone battery depletion, uncertainty propagation, and mechanism-level energy-management strategies under explicitly defined reference conditions.