基于数据挖掘技术的电池储能系统SOC状态评估

State of Charge Estimation for Battery Energy Storage System Based on Big Data Mining Technology

针对影响电池储能系统SOC运行状态的主要因素进行分析,提出了基于数据挖掘的电池储能系统SOC状态评估方法。基于粗糙集理论的数据挖掘方法,对实测的电池储能系统运行状态数据进行挖掘,提取出影响SOC的重要参数;最后利用GA-BP神经网络实现SOC状态评估,通过仿真实例验证了方法的有效性和正确性。

Combined with big data mining technology,a novel method of the SOC estimation for the battery energy storage system（ BESS） based on the rough set（ RS） and genetic algorithm-BP neural network（ GA-BP） was proposed. On the basis of the charging and discharging experiment platform of the BESS,the rough set theory was adopted to achieve the attribute reduction of the main factors related to operating status of the SOC,in order to extract the characteristic parameters properly evaluating the SOC of the BESS. The SOC estimation model of the BESS combined the RS with GA-BP neural network was established to further improve the accuracy of the SOC estimation,in which the above parameters are regarded as the input of GA-BP neural network,the operating status of the SOC as the output of the GA-BP neural network. The simulation test results verify the validity and correctness of the model.