This paper presents a data-driven approach for predicting battery calendar degradation trajectory under storage conditions, incorporating knowledge from battery aging mechanisms. The approach uses Support Vector Regression (SVR) with a novel kernel function that integrates empirical and mechanistic knowledge about battery aging. Two datasets, Dataset A and Dataset B, are used to evaluate the model's performance under different storage temperatures and State-of-Charge (SoC) levels. The results show that the SVR-based model with the battery knowledge-motivated kernel can accurately predict future battery capacity degradation under various storage conditions, outperforming other data-driven methods. The model's reliability and accuracy are validated through detailed analyses and comparisons, demonstrating its potential for improving battery health management and lifetime analysis in energy storage applications.This paper presents a data-driven approach for predicting battery calendar degradation trajectory under storage conditions, incorporating knowledge from battery aging mechanisms. The approach uses Support Vector Regression (SVR) with a novel kernel function that integrates empirical and mechanistic knowledge about battery aging. Two datasets, Dataset A and Dataset B, are used to evaluate the model's performance under different storage temperatures and State-of-Charge (SoC) levels. The results show that the SVR-based model with the battery knowledge-motivated kernel can accurately predict future battery capacity degradation under various storage conditions, outperforming other data-driven methods. The model's reliability and accuracy are validated through detailed analyses and comparisons, demonstrating its potential for improving battery health management and lifetime analysis in energy storage applications.