基于SHHO-SVR算法的锂电池剩余使用寿命预测OA

Accurate prediction of the remaining useful life(RUL)of lithium-ion batteries is essential for ensu-ring the safety of aircraft,where these batteries are widely used for their superior performance.Support vector regression(SVR)is widely applied to RUL prediction,but its performance is highly sensitive to parameter set-tings.Therefore,optimization algorithms are often introduced for parameter tuning.The commonly used Harris hawks optimization(HHO)algorithm suffers from a tendency to fall into local optima,limiting its effectiveness.To address this issue,a sparrow Harris hawks optimization(SHHO)algorithm is proposed.SHHO incorporates Skew Tent chaotic mapping,the sparrow search algorithm(SSA),multi-elite guidance,and a greedy strategy to improve global search ability and convergence accuracy.Experiments on 19 benchmark functions and the NASA lithium-ion battery dataset demonstrate that SHHO achieves better convergence and avoids local optima more effectively.The SHHO-SVR algorithm provides more accurate RUL predictions,the root mean square er-ror(RMSE)is reduced by more than 50%on average,and the prediction of the end-of-life point is more accu-rate.