State-of-charge estimation algorithm for lithium ion battery

dc.contributorAalto-yliopistofi
dc.contributorAalto Universityen
dc.contributor.advisorPavuluri, Sri Harsha
dc.contributor.authorZawadzka, Ewa
dc.contributor.schoolInsinööritieteiden korkeakoulufi
dc.contributor.supervisorSantasalo-Aarnio, Annukka
dc.date.accessioned2023-05-28T17:00:09Z
dc.date.available2023-05-28T17:00:09Z
dc.date.issued2023-05-15
dc.description.abstractThis aim of this research was to explore and develop a tool for battery State-of-Charge estimation. State-of-Charge (SOC) is the battery parameter, which serves as a crucial input information that helps to unlock battery full capability and to maintain its health state, during proper operation. Obtaining the accurate SOC value is however challenging, due to the fact that stored chemical energy cannot be directly measured. Instead SOC needs to be somehow estimated using measurable quantities. This study questions if Sigma Point Kalman Filter (SPKF) can be an optimal method for SOC estimation. Prior to developing the tool, testing hardware was prepared, which involved the calibration of Analog-to-Digital Converter (ADC) on the battery balancer board. Then the battery was subjected to the number of experiments, where it was discharged with the load profile that corresponds to its normal operation. Results were gathered for different temperatures and used to create the battery model, and later for testing and validating the SPKF model. The algorithm was first developed in the algebraic form and later translated into the computer code. Final step required tuning of the algorithm, using 3 parameters: sigma w, sigma v, sigma z0, that represents different sources of errors. The results showed that proper choice of tuning parameters was of a great importance. When chosen incorrectly, the SPKF algorithm results were not following the true value of SOC, giving no useful information. However, when tuning parameters were adjusted the algorithm was able to achieve satisfactory performance with the RMS error of around 2%. The algorithm provides not only the estimation of the SOC value, but also the information about the error bounds, providing the confidence intervals for the estimated value. This study shows that using Sigma Point Kalman Filter for estimating battery SOC provides satisfactory results and that this method is suitable for real time applications. The benefit of this method is that it doesn't needs to store significant amount of data, and it's self regulating, due to it's build-in feedback loop.en
dc.format.extent46+9
dc.identifier.urihttps://aaltodoc.aalto.fi/handle/123456789/121086
dc.identifier.urnURN:NBN:fi:aalto-202305283419
dc.language.isoenen
dc.programmeMaster’s programme in Energy Storagefi
dc.programme.majorEnergy Storagefi
dc.programme.mcodefi
dc.subject.keywordLi-ion batteryen
dc.subject.keywordSOCen
dc.subject.keywordstate of chargeen
dc.subject.keywordKalman filteren
dc.subject.keywordsigma point Kalman filteren
dc.subject.keywordreal-time applicationen
dc.titleState-of-charge estimation algorithm for lithium ion batteryen
dc.typeG2 Pro gradu, diplomityöfi
dc.type.ontasotMaster's thesisen
dc.type.ontasotDiplomityöfi
local.aalto.electroniconlyyes
local.aalto.openaccessno
Files