Design Of Battery Pack Management System For Power And Environment

The number of power and environment is increasing substantially,which causes an increasingly prominent problem how to improve security of power and environment.Therefore,monitoring and management of battery pack has become an indispensable technology.This paper proposes a new estimation method for state-of-charge of lead-acid battery,and focuses on the study and design of the software part of the battery management system.Firstly,this paper not only analyzes the working principle and charge-discharge characteristics of the lead-acid battery as well as the common influence factors of state-of-charge,but also compares a few commonly used methods of estimating state-of-charge.This paper proposes a prediction method for state-of-charge based on mathematical model and support vector machine,which produces a large number of comprehensive and reliable training samples in a short period of time by establishing the mathematical model of battery and simulating different external characteristic parameters of battery,moreover,combines VC dimension of support vector machine with structural risk minimization theory to improve the prediction accuracy of state of charge.Secondly,this paper designs and implements the battery management system based on the estimation algorithm for state-of-charge of lead-acid battery.The management system interacts data with the acquisition module by using MODBUS/TCP protocol.This system is designed as distributed architecture with arborescence-hierarchical management structure to implement collection of the battery’s feature parameters,the real-time alarm and recovery,storage and graphical visualization of data.The management system redefines prerequisite for alarm establishment and cancellation to solve the problem of frequent alarming.At last,this paper introduces building of experiment platform and testing of core functions of the management system,and analyses the test results briefly.In summary,the proposed estimation method provides a new direction for the estimation of state-of-charge.The battery pack management system designed from this method is reliable and stable,which improves security of the power and environment effectively.