Battery state estimation using high-frequency empirical model with resolved time constant

Patent metadata

Field	Value
Publication number	US-2020182937-A1
Application number	US-201816213147-A
Country	US
Kind code	A1
Filing date	Dec 7, 2018
Priority date	Dec 7, 2018
Publication date	Jun 11, 2020
Grant date	—

Abstract

An electrical system includes a battery, sensors, and a controller. The sensors output measured signals indicative of an actual state of the battery, including respective actual voltage, current, and temperature signals for each battery cell. The controller, in conducting a method, generates an estimated state of the battery, including a predicted voltage of the battery, doing so responsive to the signals using an open-circuit voltage and an output of an empirical model. An operating state of the electrical system is controlled using the estimated state. The empirical model includes low-pass/band-pass filters and a high-pass filter each with a different time-constant, the time-constants being spread over a time-constant range. Each low-pass/band-pass filter branches through a basis function(s) whose output(s) are multiplied by a respective resistance value to generate higher-frequency voltage transients. The controller sums the open-circuit voltage and voltage transients to derive the predicted voltage.

First claim

What is claimed is:

1 . An electrical system comprising:
a battery having or more battery cells; a plurality of sensors configured to output measured signals indicative of an actual state of the battery, the actual state including respective actual voltage, current, and temperature values of each of the battery cells; and a controller configured to receive the measured state signals, and responsive to the measured state signals, generate an estimated state of the battery, using at least an open-circuit voltage and an empirical model, the estimated state including a predicted voltage of the battery, and further configured to control an operating state of the electrical system in real-time responsive to the estimated state; wherein the empirical model includes a plurality of low-pass and/or band-pass filters and a high-pass filter each with a different time-constant collectively spread over a predetermined time-constant range, each of the low-pass and/or band-pass filters branching through one or more basis functions whose outputs are multiplied by a respective calibrated resistance value to generate higher-frequency voltage transients, the controller summing the higher-frequency voltage transients when deriving the predicted voltage.

2 . The electrical system of claim 1 , wherein the controller is further configured to periodically adjust the empirical model based on a difference between the predicted voltage and the actual voltage, and wherein the operating state is a charging or discharging operation of the battery.

3 . The electrical system of claim 2 , wherein the controller is configured to derive a state of charge of the battery using the estimated state, and to adjust the empirical model by periodically adjusting the respective calibrated resistances based on the state of charge and temperature.

4 . The electrical system of claim 3 , wherein the electrical system is in communication with a display device, and the controller is configured to display the state of charge via the display device.

5 . The electrical system of claim 1 , wherein the controller is further configured to generate the estimated state using a low-frequency porous electrode model in addition to the empirical model, the low-frequency porous electrode model accounting for an uneven state of charge distribution between and within opposing electrodes of each of the battery cells.

6 . The electrical system of claim 1 , wherein the controller is further configured to derive a numeric state of health of the battery using a time history of the estimated state, and to output a signal indicative of the numeric state of health.

7 . The electrical system of claim 1 , wherein the predetermined time constant range is 1 second (s) to 1000 s.

8 . The electrical system of claim 1 , wherein the empirical model uses at least three of the low-pass and/or band-pass filters.

9 . The electrical system of claim 1 , wherein at least one of the basis functions is a non-linear basis function.

10 . The electrical system of claim 1 , further comprising: an electric machine coupled to a load, such that in the discharging mode, the electric machine powers the load and in the charging mode, the electric machine extracts power from the load to recharge the battery.

11 . The electrical system of claim 10 , wherein the load is a set of drive wheels of a motor vehicle.

12 . A method for use with an electrical system having a battery with one or more battery cells, the method comprising:
measuring and outputting signals indicative of an actual state of the battery, the state signals including respective actual voltage, current, and temperature signals for each of the one or more battery cells; and responsive to the signals, generating an estimated state of the battery via a controller using at least an open-circuit voltage an empirical model, the estimated state including a predicted voltage of the battery, wherein generating the estimated state includes:
feeding the current signal through a plurality of low-pass and/or band-pass filters and a high-pass filter each having a different time-constant collectively spread over a predetermined time-constant range, each low-pass and/or band-pass filter branching through one or more basis functions;
multiplying the output of each low-pass and/or band-pass filter and the high-pass filter by a respective calibrated resistance value to generate a set of higher-frequency voltage transients; and
summing the plurality of higher-frequency voltage transients and the open-circuit voltage to derive a predicted voltage of the battery; and
controlling an operating state of the electrical system in real-time via the controller responsive to the predicted voltage.

13 . The method of claim 12 , further comprising: periodically adjusting the empirical model based on a difference between the predicted voltage and the actual voltage.

14 . The method of claim 13 , further comprising: periodically adjusting the respective calibrated resistances to adjust the empirical model.

15 . The method of claim 12 , wherein the operating state is a charging or discharging operation of the battery.

16 . The method of claim 12 , wherein the electrical system is in communication with a display device, the method further comprising: displaying the state of charge via the display device.

17 . The method of claim 13 , further comprising:
generating the estimated state using a lower-frequency porous electrode model in addition to the empirical model, the porous electrode model accounting for uneven state of charge distribution between and within opposing electrodes of each of the one or more battery cells.

18 . The method of claim 12 , the method further comprising: deriving a numeric state of health of the battery via the controller using a time history of the estimated state, and outputting a signal indicative of the numeric state of health.

19 . The method of claim 12 , wherein at least one of the basis functions is a non-linear basis function.

20 . The method of claim 12 , further comprising: powering an electric traction motor coupled to a set of road wheels of a motor vehicle in the discharging mode.

Assignees

• Gm Global Tech Operations Llc

Inventors

• Wampler Ii Charles W

Classifications

• H02J2105/37

  exchanging power with electric vehicles [EV] or with hybrid electric vehicles [HEV] · CPC title

• H02J7/865

  Battery or charger load switching, e.g. concurrent charging and load supply (H02J7/50 takes precedence) · CPC title

• H02J7/84

  Control of state of health [SOH] · CPC title

• H02J7/82 Primary

  Control of state of charge [SOC] · CPC title

• G01R31/367 Primary

  Software therefor, e.g. for battery testing using modelling or look-up tables · CPC title