基于数据驱动的动车组镍镉电池记忆效应消除策略研究  

Data-driven study on memory effect elimination strategy for Nickel-Cadmium batteries in high-speed train sets

作　　者：于天剑[1] 冯恩来 伍珣 YU Tianjian;FENG Enlai;WU Xun(School of Traffic and Transportation,Central South University,Changsha 410075,China)

机构地区：[1]中南大学交通运输工程学院,湖南长沙410075

出　　处：《铁道科学与工程学报》2024年第5期1747-1760,共14页Journal of Railway Science and Engineering

基　　金：湖南省科技人才托举工程资助项目(2022TJ-N14)。

摘　　要：在碱性镍镉蓄电池长期未彻底充放电的情况下,蓄电池会产生次级放电平台,称为“记忆效应”。该效应会导致电池可释放容量降低,需要定期消除。在动车组蓄电池检修中,传统的“记忆效应”消除方法耗时过长,会对检修工作带来很大压力。为了降低储能检修成本,对动车组镍镉电池“记忆效应”消除策略优化方案开展研究。以实际动车组三级修镍镉蓄电池电池为研究对象,通过分析电池的充放电循环中的参数特征,结合大量的电池充放电实验,提取出与电池容量密切相关的特征数据。首先,通过特征筛选和降维组合,搭建线性模型和非线性模型的交叉加权耦合方法,创建近似模型以反映电池的真实可释放容量,消除镍镉电池不一致性对优化方案实验设计带来的系统误差。其次,在此基础上,通过调整影响电池状态的高影响因子变量,构建正交实验方案。研究动车组镍镉蓄电池“记忆效应”快速、安全的消除策略。最后,通过重复验证实验和安全性检测实验对该策略进行实证,研究结果表明,提出的镍镉蓄电池“记忆效应”消除策略能够有效恢复镍镉电池真实容量,并且能够极大地缩减时间成本(约44.15%),并且有利于缓解蓄电池因过充导致电池劣化现象。该消除方案对于提升动车组镍镉电池的维护效率及确保其安全性具有重要意义。In alkaline nickel-cadmium batteries,when not subjected to full charge and discharge cycles over an extended duration,a phenomenon as“memory effect”emerges.This can result in a secondary discharge plateau which leads to diminished battery capacity,necessitating periodic rectifications.Traditional methods used to counteract this effect,particularly in the context of train battery maintenance,are labor-intensive,placing undue stress on maintenance protocols.To alleviate the burdens of energy storage maintenance,extensive research was initiated with a focus on devising optimized strategies for rectifying the memory effect in train batteries that employ nickel-cadmium cells.The article had utilized nickel-cadmium batteries sourced directly from third-level maintenance of trains.The parameters pertaining to the battery charging and discharging cycles were meticulously analyzed.Comprehensive charge-discharge experiments were undertaken,from which data that was intimately tied to battery capacity was garnered.Leveraging feature selection and dimensionality reduction techniques,a novel modeling approach was developed.It termed the cross-weighted coupling method,which amalgamated both linear and non-linear models,yielding an approximate model that mirrors the true dischargeable capacity of the battery.This method effectively eliminated the systemic errors inherent in experimental designs,which were often due to inconsistencies in nickel-cadmium batteries.Building upon this foundational work,by finely tuning critical factors that impact the battery’s state,an orthogonal experimental methodology was conceptualized with its core aim being rapidly and safely mitigate the memory effect prevalent in train nickel-cadmium batteries.Rigorous validation coupled with exhaustive safety assessments affirmed the efficacy of this new approach.It not only revives the true capacity of the nickel-cadmium battery but also reduces the time expenditure by approximately 44.15%.Additionally,it curtails battery degradation stemming from ov

关 键 词：动车组 镍镉蓄电池 记忆效应 信息挖掘 交叉融合建模方法 正交实验 

分 类 号：TM912.2[电气工程—电力电子与电力传动]