基于注意力增强Uniformer的锂电池剩余使用寿命预测

打开文本图片集

主题词：锂电池剩余使用寿命预测数据驱动统一变形器注意力引导机制坐标注意力中图分类号：TM912 文献标志码：A DOI：10.19620/j.cnki.1000-3703.20240396

Remaining Useful Life Prediction of Lithium Battery Based on Attention Enhancement Uniformer

LiaoLiefa²，LiuYingbao¹，ZhanYumin1 （1.SchoolofInformationEngineering，Jiangxi UniversityofScienceandTechnology，Ganzhou34ooo;2.Jiangxi Moder Polytechnic College，Nanchang 330095）

【Abstract】To address the isse of dynamic changes in data and limited aging data inthe Remaining Useful Life （RUL） predictionof lithium-ion bateries，thispaperproposes theRUL predictionmodelof AttentionEnhancementUniformer （AEUniformer）torealizecomprehensiveinformationperceptionbyintegratingtheadvantagesofConvolutionalNeuralNetwork （CNN）andSelf-Atention Mechanism through Uniformer.Attention Guiding Mechanism （AGM）and CoordAttentionare designedtorealize powerfulfeature extraction.ExperimentalresultsshowthatAEUniformercanachieveacurateand fastRUL prediction withonlyasingle agingcycle，andthe MAPE prediction errorsofthe datasets are2.7%and 6.16%，respectively， demonstrating the accuracy of the method.

KeyWords：Lithium-ion battery，Remaining Useful Life （RUL），Data-driven，Uniformer. AttentionGuidingMechanism （AGM），CoordAttention

【引用格式】廖列法，刘映宝，占玉敏.基于注意力增强Uniformer的锂电池剩余使用寿命预测[J].汽车技术，2025（6）：36-44.LIAOLF，LIUYB，ZHANYM.Remaining UsefulLifePredictionofLithium BatteryBasedonAttentionEnhancementUniformer[J].Automobile Technology，2025（6）： 36-44.

1前言

锂电池因其高能量密度、长循环寿命和低自放电率等特点而被广泛应用于储能和供电领域[。（剩余12574字）