基于改进的Bi-LSTM水电站蓄电池状态监测研究  

Research on Multi-module Cooperative Blocking Technology for Hydropower Station Battery Systems

作　　者：方建文 邬建春 秦谱 郭夏明 罗荣雄 FANG Jianwen;WU Jianchun;QIN Pu;GUO Xiaming;LUO Rongxiong

机构地区：[1]云南大唐国际那兰水电开发有限公司,云南红河661100

出　　处：《电力系统装备》2025年第2期94-96,共3页Electric Power System Equipment

摘　　要：蓄电池在水电站直流系统中的健康状态至关重要,其工作状态不仅影响电站的稳定运行,还决定了故障预防和维护的效率。蓄电池的充电状态、健康状态及故障诊断是确保蓄电池可靠性和寿命管理的核心指标。为了有效监测蓄电池的SOC、SOH及其故障特征,文章提出了一种基于改进的双向长短期记忆网络的蓄电池状态监测模型。该模型能够精准捕捉蓄电池在不同运行工况下的动态变化,归纳出各特征参数的变化规律,为蓄电池的状态监测与故障诊断提供了有效的解决方案。The health status of batteries in the DC system of hydropower stations is crucial.Their working condition not only affects the stable operation of the power station,but also determines the efficiency of fault prevention and maintenance.The charging status,health status,and fault diagnosis of batteries are the core indicators to ensure their reliability and lifespan management.In order to effectively monitor the SOC,SOH,and fault characteristics of batteries,this article proposes a battery state monitoring model based on an improved bidirectional long short-term memory network.This model can accurately capture the dynamic changes of the battery under different operating conditions,summarize the variation laws of various characteristic parameters,and provide an effective solution for the state monitoring and fault diagnosis of the battery.

关 键 词：铅酸蓄电池 Bi–Lstm SOC SOH 故障检测 

分 类 号：TM912[电气工程—电力电子与电力传动]