【摘要】 综述了动力电池荷电状态(SOC)估算的传统方法、人工智能方法的原理,分析了各估算方法的优缺点,给出了其他SOC估算的实现策略,如自适应卡尔曼滤波法、主元分析法以及遗传算法(GA)-BP神经网络法。研究表明,在动力电池SOC估算的实际应用中,要充分考虑实测数据、软硬件条件来选择相应的动力电池模型,综合考虑各种SOC估算策略,以此来提高SOC估算的精度。更多还原

【Abstract】 The principles of the traditional method and artificial intelligence method of power battery SOC estimation were reviewed, and their advantages and disadvantages were analyzed. The other SOC estimation strategies were given, such as adaptive Kalman filter method, principal component analysis and genetic algorithm(GA)-BP neural network. The results show that the actual data and software and hardware conditions should be fully considered to select the appropriate battery model in the application of battery SOC estimation. All kinds of SOC estimation strategies should be fully considered to improve the accuracy of SOC estimating.更多还原