← 返回

基于自适应无迹卡尔曼滤波与支持向量机的锂聚合物电池荷电状态估计

Lithium Polymer Battery State-of-Charge Estimation Based on Adaptive Unscented Kalman Filter and Support Vector Machine

语言:

中文摘要

本文提出了一种结合自适应无迹卡尔曼滤波(AUKF)与最小二乘支持向量机(LSSVM)的锂聚合物电池荷电状态(SOC)估计算法。通过引入移动窗口法,在初始训练样本有限的情况下,有效建立了高精度的电池模型,提升了SOC估计的准确性与鲁棒性。

English Abstract

An accurate algorithm for lithium polymer battery state-of-charge (SOC) estimation is proposed based on adaptive unscented Kalman filters (AUKF) and least-square support vector machines (LSSVM). A novel approach using the moving window method is applied, with AUKF and LSSVM to accurately establish the battery model with limited initial training samples. The effectiveness of the moving window model...
S

SunView 深度解读

该技术对阳光电源的储能业务(如PowerTitan、PowerStack及ST系列PCS)具有极高价值。SOC估计是BMS的核心算法,直接影响储能系统的可用容量、充放电策略及安全性。通过引入AUKF与LSSVM算法,可显著提升在复杂工况(如温度波动、老化)下的SOC估算精度,减少因估算偏差导致的电池过充过放风险。建议研发团队将其集成至iSolarCloud智能运维平台,利用大数据训练模型,进一步优化储能电站的调峰调频性能,提升系统全生命周期的经济效益。