Application of Pseudo Random Binary Sequences for Battery Impedance Estimation via Motor Drive Systems
Translated title
Anvendelse af pseudo tilfældig binære sekvenser til estimering af batteri impedans i motordrevne systemer
Author
Bjerre, Marcus Hoegh
Term
4. term
Education
Publication year
2025
Submitted on
2025-05-28
Pages
26
Abstract
Dette projekt undersøger, hvordan Pseudo Random Binary Sequence (PRBS) kan bruges til at estimere batteriets impedans i realtid i et system, hvor et batteri forsyner en motorstyring. Baggrunden er, at traditionelle metoder som elektro-kemisk impedansspektroskopi (EIS) er præcise, men uhensigtsmæssige til online overvågning på grund af tidsforbrug og kompleksitet. Metodisk implementeres PRBS som en forstyrrelse i iq-strømreferencen i en Field Oriented Control (FOC)-struktur, der driver en Permanent Magnet Synchronous Machine (PMSM). Batteriets spænding og strøm registreres og behandles i frekvensdomænet for at udlede impedansen. En 4-RC batterimodel, parametriseret med reference fra EIS (bl.a. via Nyquist-halvcirkler), bruges i simuleringer til at vurdere, hvordan PRBS’ bit-længde, clockfrekvens og amplitude påvirker kvaliteten af impedansestimatet. Resultaterne viser klare trade-offs: længere sekvenser forbedrer lavfrekvent opløsning (diffusionsområdet), men øger testtid og følsomhed for støj; højere clockfrekvens forbedrer højfrekvent registrering, men forringer lavfrekvent estimat. Studiet giver designretningslinjer for valg af PRBS-parametre afhængigt af målte frekvensområder og beregningsressourcer. Da arbejdet er simuleringsbaseret, fremhæves behovet for eksperimentel validering samt opmærksomhed på målestøj og korrekt amplitudeskalering ved praktisk implementering.
This project investigates the use of Pseudo Random Binary Sequences (PRBS) for real-time battery impedance estimation in systems where a battery powers a motor drive. The motivation is that conventional methods such as Electrochemical Impedance Spectroscopy (EIS) are accurate but impractical for online monitoring due to their duration and complexity. The approach injects a PRBS disturbance into the iq current reference of a Field Oriented Control (FOC) scheme driving a Permanent Magnet Synchronous Machine (PMSM). Battery-side voltage and current are recorded and processed in the frequency domain to extract impedance. A 4-RC battery model, parameterized using EIS data (including Nyquist semicircle features), is used in simulations to evaluate how PRBS bit length, clock frequency, and amplitude affect estimation quality. The results reveal trade-offs: longer sequences improve low-frequency resolution (diffusion region) but increase test time and susceptibility to noise; higher clock frequencies enhance high-frequency capture but degrade low-frequency estimation. The study offers design guidelines for selecting PRBS parameters based on target frequency ranges and computational constraints. As the work is simulation-based, future efforts should focus on experimental implementation and validation, with attention to measurement noise and appropriate amplitude scaling.
[This abstract was generated with the help of AI]
Documents