Assessment and Analysis of Electric Capacity for Lithium Batteries Using Neural Networks

• Main Author: 彭義淵
• Other Authors: 曾仲熙
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/43554905630276786596

碩士 === 明新科技大學 === 電機工程研究所 === 99 === In this thesis, the topic of electric capacity for lithium battery is studied . The purpose of this these is to manage the battery charge status (state of charge, SOC). Through the proposed measurement system, the status of battery charge is estimated. As accurate estimation of SOC can be obtained only by fully discharging the battery. This procedure is impractical because large amount of time is required. In this thesis, it is assumed that SOC decreases linearly with time if discharge occurs with a periodical current profile. First, Matlab control algorithm for charging and discharging is performed using acquisition card (PCI1711) with Matlab Simulink, then data of extracted charge, internal resistance and no-load voltage are all collected and save，finally these data are used to build up the SOC model for the lithium battery using neural network. After training, the weightings and the bias rule of neural network can be obtained and be used to build up the SOC model of lithium battery. Finagling, the SOC model of neural network can be used to determine the SOC of a new battery. By the experimental results, the feasibility and accuracy of the proposed method can be verified and the estimation of the battery capacity mean be achieved. By the proposed method, SOC of a new battery can be determined immediately using the measurements of extracted charge, internal resistance, and no-load voltage for one-cycle of charging and discharging profile only.