基于高斯过程回归的电池容量预测模型

Battery capacity prediction model based on Gaussian process regression

为了更好地对电池健康状况和使用寿命进行预测,为装备保养维修提供更为详尽的决策依据,针对电池动态、时变、非线性的系统特点,提出了一种基于高斯过程回归的电池容量预测模型。基于不同的核函数对电池容量进行回归预测,同时对比了灰色模型和神经网络模型的预测效果。仿真结果表明:高斯过程回归模型的预测效果要优于其他模型的预测效果;对于电池容量的预测,平方指数协方差函数和二次有理协方差函数的组合模型可以取得良好的预测结果,预测误差下降了31.157%。

In order to better predict the health and remaining life of the battery,and provide more detailed decision-making basis for equipment maintenance and repair,it proposes a battery capacity prediction model based on Gaussian process regression for the characteristics of battery dynamics,time-varying and nonlinear system.Regression prediction of battery capacity is based on different kernel functions,and compare the prediction effects of gray model and neural network model.The simulation results show that the prediction effect of Gaussian process regression model is better than that of other models.For the prediction of battery capacity,the combined model of square exponential covariance function and quadratic rational covariance function can obtain good prediction results and prediction error.It fell by 31.157%.