# US 12,601,787 B2 · Estimation of battery equivalent circuit model parameters by decomposition of sense current and terminal voltage into subbands > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,601,787 B2 - Document ID: US12601787B2 - Status: Granted - Kind: B2 - Application number: 18/308,449 - Filed: 2023-04-27 - Published: 2024-04-25 - Granted: 2026-04-14 - Earliest claimed: 2022-10-12 - Assignees: Cirrus Logic, Inc. - Applicants: Cirrus Logic International Semiconductor Ltd. - Inventors: Samuel P. Ebenezer; Emmanuel A. Marchais; Eric J. King - Web page: https://patentgrep.com/patent/US12601787B2 ## CPC - G01R 31/367: Software therefor, e.g. for battery testing using modelling or look-up tables - G01R 31/3842: combining voltage and current measurements - G01R 31/389: Measuring internal impedance, internal conductance or related variables - G01R: MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES (indicating correct tuning of resonant circuits H03J3/12) ## Abstract A method for estimating equivalent circuit model parameters of a battery may include measuring a battery voltage across terminals of the battery and a battery current drawn from the battery, decomposing the battery voltage and the battery current into a plurality of sub-bands, each sub-band of the plurality of sub-bands based on a time constant that characterizes a temporal behavior of the battery, for each sub-band of the plurality of sub-bands, estimating an equivalent resistance for such sub-band based on a spectral content of the battery voltage and battery current for such sub-band, and estimating an open circuit voltage of the battery based at least on the spectral content of the battery voltage and battery current present in one of the plurality of sub-bands and the equivalent resistances of the plurality of sub-bands. ## Claims ### Claim 1 A method for estimating equivalent circuit model parameters of a battery, comprising: measuring a battery voltage across terminals of the battery and a battery current drawn from the battery; decimating, via a multi-rate filter bank comprising a plurality of decimators, the battery voltage and battery current into a plurality of low-frequency signals, wherein the plurality of decimators are communicatively coupled to a plurality of filters configured to receive the plurality of low-frequency signals and generate a plurality of sub-bands, each sub-band of the plurality of sub-bands based on a time constant that characterizes a temporal behavior of the battery; selecting, via a signal selector communicatively coupled to outputs of the plurality of filters and to inputs of a plurality of transform-resistor pairs, from the plurality of sub-bands, a sub-band for each transform-resistor pair based on an associated time constant; estimating, via the transform-resistor pairs, an equivalent resistance for each sub-band of the plurality of sub-bands based on a spectral content of the battery voltage and battery current for such sub-band; and estimating, via an open circuit voltage tracking module, an open circuit voltage of the battery based at least on the spectral content of the battery voltage and battery current present in one of the plurality of sub-bands and the equivalent resistances of the plurality of sub-bands. ### Claim 2 (depends on 1) The method of claim 1, wherein the time constants are fixed. ### Claim 3 (depends on 1) The method of claim 1, wherein the time constants are adaptive. ### Claim 4 (depends on 1) The method of claim 1, wherein estimating the equivalent resistance for each sub-band comprises adapting such equivalent resistance using an adaptive algorithm comprises one of an adaptive recursive-least squares, total-least squares, normalized least-mean-squares, total least squares cost function, or gradient descent-based algorithm. ### Claim 5 (depends on 4) The method of claim 4, further comprising updating the equivalent resistance for each sub-band only when a root-mean-square level of the battery current is above a threshold. ### Claim 6 (depends on 4) The method of claim 4, further comprising updating the equivalent resistance for each sub-band to minimize an error for such sub-band. ### Claim 7 (depends on 4) The method of claim 4, further comprising updating the equivalent resistance for each sub-band to minimize a full-band error. ### Claim 8 (depends on 4) The method of claim 4, wherein a learning rate of the adaptive algorithm is based on a root-mean-square level of current within the sub-band. ### Claim 9 (depends on 8) The method of claim 8, wherein the adaptive algorithm comprises an update step that updates the equivalent resistance based on an error. ### Claim 10 (depends on 4) The method of claim 4, wherein a regularization step is applied during a parameter update step of the adaptive algorithm. ### Claim 11 (depends on 1) The method of claim 1, further comprising: determining a spectral sufficiency of the battery current in each of the sub-bands; and generating an augmented current to be drawn from the battery when a spectral insufficiency is determined to exist in one or more sub-bands. ### Claim 12 (depends on 1) The method of claim 1, further comprising estimating an inductance of the equivalent circuit model based at least on the spectral content of the battery voltage and battery current present in one of the plurality of sub-bands. ### Claim 13 A system for estimating equivalent circuit model parameters of a battery, comprising circuitry for measuring a battery voltage across terminals of the battery and a battery current drawn from the battery, wherein the circuitry comprises: a multi-rate filter bank comprising a plurality of decimators configured to decimate the battery voltage and battery current into a plurality of low-frequency signals, wherein the plurality of decimators are communicatively coupled to a plurality of filters configured to receive the plurality of low-frequency signals and generate a plurality of sub-bands, each sub-band of the plurality of sub-bands based on a time constant that characterizes a temporal behavior of the battery; a signal selector communicatively coupled to outputs of the plurality of filters and to inputs of a plurality of transform-resistor pairs, wherein the signal selector is configured to receive the plurality of sub-bands and select, for each transform-resistor pair, a sub-band based on an associated time constant, wherein the plurality of transform-resistor pairs are configured to estimate an equivalent resistance for each sub-band of the plurality of sub-bands based on a spectral content of the battery voltage and battery current for such sub-band; and an open circuit voltage tracking module configured to estimate an open circuit voltage of the battery based at least on the spectral content of the battery voltage and battery current present in one of the plurality of sub-bands and the equivalent resistances of the plurality of sub-bands. ### Claim 14 (depends on 13) The system of claim 13, wherein the time constants are fixed. ### Claim 15 (depends on 13) The system of claim 13, wherein the time constants are adaptive. ### Claim 16 (depends on 13) The system of claim 13, wherein estimating the equivalent resistance for each sub-band comprises adapting such equivalent resistance using an adaptive algorithm comprising one of an adaptive recursive-least squares, total-least squares, normalized least-mean-squares, total least squares cost function, or gradient descent-based algorithm. ### Claim 17 (depends on 16) The system of claim 16, the circuitry further configured for updating the equivalent resistance for each sub-band only when a root-mean-square level of the battery current is above a threshold. ### Claim 18 (depends on 17) The system of claim 17, the circuitry further configured for updating the equivalent resistance for each sub-band to minimize a full-band error. ### Claim 19 (depends on 16) The system of claim 16, the circuitry further configured for updating the equivalent resistance for each sub-band to minimize an error for such sub-band. ### Claim 20 (depends on 16) The system of claim 16, wherein a learning rate of the adaptive algorithm is based on a root-mean-square level of current within the sub-band. ### Claim 21 (depends on 20) The system of claim 20, wherein the adaptive algorithm comprises an update step that updates the equivalent resistance based on an error. ### Claim 22 (depends on 16) The system of claim 16, the circuitry further configured for applying a regularization step during a parameter update step of the adaptive algorithm. ### Claim 23 (depends on 13) The system of claim 13, the circuitry further configured for: determining a spectral sufficiency of the battery current in each of the sub-bands; and generating an augmented current to be drawn from the battery when a spectral insufficiency is determined to exist in one or more sub-bands. ### Claim 24 (depends on 13) The system of claim 13, the circuitry further configured for estimating an inductance of the equivalent circuit model based at least on the spectral content of the battery voltage and battery current present in one of the plurality of sub-bands.