基于数字孪生的煤矿机器人锂电池SOC评估方法

Study on SOC Evaluation Method of Lithium Battery for Coal Mine Robot Based on Digital Twins

针对煤矿机器人锂电池的荷电状态(SOC)实时监控困难和安全管理问题,提出了一种基于数字孪生的煤矿机器人锂电池SOC估计模型.首先,搭建了基于数字孪生的煤矿机器人锂电池SOC估计模型;其次,将锂电池充电时的电压、电流作为数字孪生模型的模拟参数,利用遗传算法优化BP神经网络模型,建立并优化锂电池充电时的电压、电流与SOC之间的关系,实现对数字孪生模型的更新及煤矿机器人锂电池数字孪生模型的SOC估计;最后,对两种不同的锂电池进行SOC估计实验,同时对比传统BP神经网络与基于GA-BP的数字孪生模型实验结果,发现所提模型模拟得到的SOC估计值明显优于传统BP神经网络,实验结果误差小,表明了该方法的有效性,对于煤矿机器人锂电池的科学管理具有重要意义.

Aiming at the real-time monitoring difficulties and safety management problems of lithium battery in coal mine robot,a SOC estimation model of lithium battery in coal mine robot based on digital twins is proposed.Firstly,the SOC estimation model of lithium battery of coal mine robot based on digital twins is built.Secondly,according to the voltage and current of lithium battery charging as the simulation parameters of digital twin model,the BP neural network model is optimized by genetic algorithm,and the relationship between voltage,current and SOC of lithium battery charging is established and optimized,so as to realize the update of digital twin model and the SOC estimation of digital twin model of lithium battery of coal mine robot.Finally,two different lithium battery SOC estimation experiments are carried out,and the experimental results of the traditional BP neural network and ga-BP digital twin model are compared.The SOC estimation obtained by the digital twin model is obviously better than that obtained by the traditional BP neural network,and the error of the experimental results is small,indicating the effectiveness of the method.It is of great significance to the scientific management of lithium battery for coal mine robot.