计及关键机理特征的锂离子电池多层次SOC/SOH/SOP协同估计是提高插电式混合动力汽车安全性和社会接受度的一个最新发展方向，也是最具前景的技术路线之一。本课题拟围绕这一前沿技术所涉及的机理模型参数敏感度分析与简化、高精度鲁棒多尺度SOC/SOH估计算法构建、感知多变量的SOP自适应估计策略设计三个关键科学问题，采用理论分析、建模仿真、算法模拟及实验验证相结合的方法开展研究。分析锂离子电池机理模型参数敏感度，构建模型简化的数学框架，对比研究不同模型降阶方法，探索新型模型结构，以获取精度与计算效率之间的理想（最佳）折中；结合简化机理模型和在线优化的优势，设计新的多尺度状态估计算法，提高SOC/SOH估计的精度和鲁棒性，开展充分的实验验证；构建感知多变量信息的SOP自适应估计及实验验证方案，提升峰值功率估算的精度与环境感知能力。研究结果将丰富和发展插电式混合动力汽车整车能量管理及电池管理技术。

Critical mechanistic characteristics-aware hierarchical co-estimation of Li-ion battery SOC/SOH/SOP is arguably a novel development direction and a promising technology for significantly improving safety and public acceptance of plug-in hybrid electric vehicles. This project will concentrate on three critical scientific questions/challenges in such a cutting-edge technology: parameter sensitivity analysis and simplification of physics-based electrochemical models, robust high-accuracy multi-scale SOC/SOH estimation, and multi-variable-conscious adaptive SOP indication. Theoretic analysis, modeling/emulation, algorithm simulation, and experimental validation will be jointly leveraged to advance the project. The research will analyze parameter sensitivity of Li-ion battery electrochemical models, formulate mathematical framework of model reduction, and compare different simplifying methods, in order to explore new model structures with desirable (optimum) balance between accuracy and computational efficiency. In light of salient advantages of physics-based models and online (moving horizon) optimization, innovative multi-scale SOC/SOH estimation algorithms will be synthesized to increase estimation precision and resilience. Systematic comparisons with existing estimation schemes will also be made. With the intention to increase the accuracy and environmental consciousness of peak power evaluation, multi-variable-conscious adaptive SOP estimation and experimental validation schemes will be mathematically formulated and solved, where comparative studies and experimental validation are sufficiently performed. Research results will do enrich and even transform vehicle-level energy management and battery management technologies for plug-in hybrid electric vehicles.