基于虚拟传感器的电池系统温度预测方法研究  

Study on Temperature Prediction Method of Battery System Based on Virtual Sensor

作　　者：董红赞 董瑢 陈欣 DONG Hongzan;DONG Rong;CHEN Xin(Shanghai Electrical Tools Research Institute(Group)Co.LTD,Shanghai 200233,China)

机构地区：[1]上海电动工具研究所(集团)有限公司,上海200233

出　　处：《集成电路应用》2022年第6期32-35,共4页Application of IC

摘　　要：阐述一种基于虚拟传感器的电池系统温度预测方法,建立基于电池充放电发热及冷却机理的电池仿真模型,实现对于不同环境、工况下电池温升的模拟,采集电池系统充放电过程中温升变化数据,验证、改进仿真模拟模型,基于仿真数据间接获取虚拟传感器的输出结果,最后融合测量温度和虚拟被测温度,提取关键特征参数,结合深度神经网络建立温升动态预测模型,实现对于工况变化后电池系统温升的快速预测。该方法适用于预测电池系统充放电过程中的温度变化,通过融合测量、虚拟测量数据建构预测模型,从而在低成本前提下实现对于电池温度变化的预测。This paper discloses a temperature prediction method of battery system based on virtual sensor, including: battery simulation model is established based on the charging and discharging heating and cooling mechanism of battery, to realize the simulation of battery temperature rise in different environments and working conditions;The temperature rise data of the battery system during charging and discharging were collected, the simulation model was verified and improved, and the output results of the virtual sensor were obtained indirectly based on the simulation data. The key characteristic parameters were extracted by integrating the measured temperature and the virtual measured temperature, and the dynamic prediction model of temperature rise was established by combining the deep neural network to realize the rapid prediction of the temperature rise of the battery system after the change of working conditions. This method can be used to predict the temperature change of battery system during charging and discharging process. The prediction model is constructed by integrating measurement data and virtual measurement data, and the prediction of battery temperature change can be realized at low cost.

关 键 词：虚拟传感器 温度预测 电池系统 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置] TB34[自动化与计算机技术—控制科学与工程]