Font Size: a A A

Study On Discharge Characteristics And Capacity Estimation Of Lithium Battery At Low Temperature

Posted on:2022-12-11Degree:MasterType:Thesis
Country:ChinaCandidate:R LiuFull Text:PDF
GTID:2492306758469714Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
With the continuous development of science and society,lithium batteries with rechargeable,recyclable and environmentally friendly attracts more and more attention,and are being used in new energy automobiles,energy storage systems,small electronic products,aerospace and other industries.However,because the performance of lithium batteries is seriously affected by temperature,its low temperature performance has become a major factor limiting the further application of lithium batteries.Therefore,it is very important to study the discharge performance of lithium batteries under low temperature,accurately estimate the charging status of lithium batteries under different low temperature conditions,and to propose a better way to use lithium batteries at low temperature.This paper mainly studies the following:(1)Based on the working principle of lithium batteries,chemical reaction process and physical structure of lithium batteries,the development history of lithium batteries,low temperature characteristics and commonly used algorithms for estimating the charging status of lithium batteries are described in detail.(2)Using lithium cobalt and nickel-cobalt-manganese ternary as composite cathode materials and lithium-ion batteries with negative lithium titanate as experimental objects,the low-temperature discharge characteristics of lithium batteries were designed from the point of view that environmental temperature and working current have great influence on the performance of lithium batteries.The experiments of low temperature discharge characteristics mainly include:activation experiments of lithium batteries,standard capacity and energy tests,low temperature discharge capacity and energy tests,low temperature open circuit voltage tests,low temperature internal resistance tests,low temperature small rate charge and discharge tests.According to the low-temperature discharge data obtained from the above experiments,analyze the low-temperature discharge characteristics of lithium battery and establish a database for the low-temperature state of charge estimation algorithm of lithium battery.(3)The experimental data show that the discharge performance of lithium batteries is greatly affected by low temperature environment.With the decrease of discharge temperature,the discharge capacity and energy decrease to some extent,while the ohm internal resistance and polarization internal resistance increase to some extent.Under the same charging state,the open-circuit voltage decreases with the decrease of temperature,and the decrease trend is accelerated.However,under different temperature conditions,the overall decrease trend of the curve is similar.The capacity increment analysis and differential voltage analysis show that the effect of positive and negative activating materials and lithium ions participating in the reaction on battery performance varies in different temperature ranges.(4)BP neural network and LSTM neural network are used to estimate the charging state of lithium batteries at different low temperatures and different discharge rates,respectively.By comparing and analyzing the estimation errors of the two algorithms under the same test data,the maximum absolute error of the BP network is 0.0241,the average RMSE is 0.0089,the average MAE is 0.0075,and the average~2 is 0.9987.The maximum absolute estimation error of LSTM neural network is 0.0132,the average RMSE is 0.0068,the average MAE is0.0056,and the average~2 is 0.9992.Therefore,LSTM neural network is more advantageous in estimating the charging status of lithium batteries under the influence of both environmental temperature and discharge rate.
Keywords/Search Tags:Lithium-ion battery, Low temperature discharge characteristics, Lithium battery SOC, BP network, LSTM neural network
PDF Full Text Request
Related items