微型电脑应用

为了准确估计动力电池的SOC(state of charge)值，研究一种基于改进长短期记忆(LSTM)的新能源汽车动力电池SOC联合估计方法。分析不同因素对新能源汽车动力电池SOC的影响。基于此，构建基于卷积神经网络(CNN)-LSTM估计模型，以实现电池SOC联合估计。结果表明，改进后的LSTM训练均方误差更小，说明改进后LSTM拟合程度更好，具有实用性。

To accurately estimate the SOC(state of charge) value of power batteries, a SOC joint estimation method for new energy vehicle power batteries based on improved long short-term memory(LSTM) is studied. The paper analyzes the impact of different factors on the SOC of new energy vehicle power batteries. Based on this, a convolutional neural network(CNN)-LSTM estimation model is constructed to achieve joint estimation of battery SOC. The results show that the improved LSTM training has a smaller mean square error, indicating a better fitting degree and practicality of the improved LSTM.