直流供电系统铅酸蓄电池组性能诊断方法

论文总字数：23751字

摘 要

　 阀控式密封铅酸蓄电池(Valve Regulated Lead Acid Battery)作为一种常规的储能装置，广泛地应用于我们生活中的各个重要领域。电池组损坏或者退役主要是因为一组电池中存在少量的劣化单体，找到并替换掉劣化单体，电池组可以继续服役，从而延长了电池组的使用寿命。此外，及时了解电池剩余容量，对于保证供电系统的安全可靠性具有重要的意义，而且可以降低企业管理成本。

本文首先对国内外蓄电池研究现状进行综述，其次对蓄电池的结构原理以及相关的特征参数研究，利用在线监测系统测量实验数据，绘制放电曲线并深入研究电池放电过程。然后使用模糊聚类方法，利用单体电池电压之间的关系，建立了基于模糊聚类的劣化单体电池检测算法并进行了实验验证，结果表明算法可以对电池健康状态进行正确分类。

本文还分析比较传统了荷电态算法的优缺点，建立了蓄电池荷电态BP神经网络预测算法并进行了仿真研究，结果表明，算法能够准确预测电池荷电态，最大误差低于1%。

关键词：高压直流供电系统；阀控式密封铅酸蓄电池；模糊聚类分析算法;电池组健康状态、电池组荷电态；BP神经网络

Abstract

Valve Regulated Lead Acid Battery, as a conventional storage device, is widely used in many important fields of our life. The battery pack is damaged or decommissioned mainly because there is a small amount of deteriorated monomer in a battery pack. If the deteriorated monomer is found and replaced, the battery pack can continue to serve, thus extending the service life of the battery pack. In addition, timely understanding of the residual capacity of the battery is of great significance to ensure the safety and reliability of the power supply system, and can reduce enterprise management costs.

In this paper, firstly, the research status of battery at home and abroad is summarized. Secondly, the structure principle of battery and related characteristic parameters are studied. By using online monitoring system to measure experimental data, the discharge curve is drawn and the battery discharge process is deeply studied. Then, the fuzzy clustering method is used to establish the detection algorithm of degraded single battery based on fuzzy clustering based on the relationship between the voltage of single battery.

This paper also analyzes and compares the advantages and disadvantages of the traditional algorithm of SOC, establishes the prediction algorithm of the battery charged state BP neural network, and carries on the simulation research, the results show that the algorithm can accurately predict the battery charged state, the maximum error is less than 1%.

Keywords: HVDC power supply system，VRLA battery, fuzzy clustering analysis algorithm, state of charge, BP neural network

目 录

摘要 3

Abstract 4

目 录 5

第一章 绪论 1

1.1研究背景 1

1.2国内外研究现状 2

1.2.1研究现状 2

1.2.2落后电池检测研究现状 3

1.2.3 剩余荷电态预测 3

1.3论文组织结构 4

第二章 VRLA蓄电池工作原理及失效分析 5

2.1VRLA蓄电池工作原理 5

2.2VRLA蓄电池常见失效原因分析 5

2.2.1蓄电池失水 5

2.2.2负极硫酸盐化 6

2.2.3正极板腐蚀 6

2.2.4热失控 6

2.3本章小结 7

第三章 VRLA蓄电池放电特征曲线和参数 8

3.1VRLA蓄电池放电特征曲线 8

3.2VRLA蓄电池特征参数 8

3.2.1VRLA蓄电池电压 8

3.2.2VRLA蓄电池电流 9

3.2.3VRLA蓄电池温度 9

3.2.4VRLA蓄电池内阻 10

3.3本章小结 11

第四章 VRLA蓄电池组健康状态(SOH)分类 12

4.1概述 12

4.2模糊聚类分析检测电池组劣化单体算法原理及实现步骤 12

4.2.1模糊聚类分析算法基本原理 12

4.2.2模糊聚类分析检测算法实现步骤 12

4.3模糊聚类分析算法对蓄电池组健康状态检测结果 13

4.3.1 待测样本初始值 13

4.3.2 VRLA蓄电池健康分类结果 14

4.4本章小结 15

第五章 VRLA蓄电池剩余荷电态(SOC)估测 16

5.1概述 16

5.1.1荷电态（SOC）定义 16

5.2.1传统荷电态预测方法 16

5.2BP神经网络算法 17

5.2.1BP神经网络算法原理 17

5.2.2BP神经网络算法具体步骤 19

5.3 BP神经网络算法对蓄电池组剩余荷电态预测结果 20

5.4本章小结 23

第六章 总结与展望 24

参考文献 25

致 谢 27

第一章 绪论

1.1 研究背景

剩余内容已隐藏，请支付后下载全文，论文总字数：23751字