To tackle low efficiency and poor cross-condition universality in traditional constant-current experiments for predicting lead-acid batteries’ remaining discharge time, this study proposes a least squares-based exponential modeling approach. A universal discharge prediction model for 20A–100A is developed, using segmented functions: quadratic relationships for low currents (20A < I ≤ 50A) to reflect nonlinear electrochemical self-catalysis, and linear modeling for high currents (50A < I ≤ 100A) per Fick's diffusion control theory. Experimental results validate the robustness of the proposed model, with mean relative errors remaining within 5.48% across all tested currents. Notably, the prediction curve for the typical 55A case exhibits a high degree of consistency with actual discharge trends, demonstrating the model's accuracy and reliability. The innovation of this study lies in the development of a universal discharge prediction model that combines exponential functions with segmented current relationships, providing a more accurate and efficient solution for battery discharge prediction. Future research directions will focus on improving the model's adaptability to varying temperatures, refining the correction mechanisms for variable-current conditions, and integrating battery health-state assessments to further enhance the universality and applicability of the model in diverse industrial scenarios.
