Model-based capacity and resistance correction for rechargeable battery fuel gauging

Abstract

In some examples, a host device includes a battery capacity and / or battery resistance prediction model. The host device may predict a battery capacity value and / or battery resistance value of a rechargeable battery and compare the predicted battery capacity and / or predicted battery resistance with a battery capacity value and / or battery resistance value stored in a fuel gauge. If the difference is greater than a maximum error threshold, the host device may overwrite the battery capacity value and / or battery resistance value stored in the fuel gauge with the predicted battery capacity value and / or predicted battery resistance value.

Description

Technical Field

[0001] The present disclosure relates to batteries, and more particularly, to calibrating the capacity and resistance values of rechargeable batteries. Background Art

[0002] Many devices, including implantable medical devices, laptops, tablet computers, and cell phones, utilize rechargeable batteries. These devices also typically include a battery management system that provides a graphical representation of the battery's status to the user of the device, such as how fully the battery can be charged. Summary of the Invention

[0003] In some aspects, the present disclosure relates to a host device that utilizes a rechargeable battery, such as, for example, an implantable medical device, a laptop computer, a tablet computer, or a cellular telephone.

[0004] In one example, the present disclosure relates to a host device comprising a memory; and one or more processors coupled to the memory, the one or more processors configured to: predict at least one of a battery capacity value or a battery resistance value of a rechargeable battery each time, the rechargeable battery powering the host device; calculate a difference between at least one of the predicted battery capacity value or the predicted battery resistance value and at least one of the battery capacity value or the battery resistance value stored in a memory of a fuel gauge; determine whether the difference exceeds a predetermined maximum error threshold; and based on the difference exceeding the predetermined maximum error threshold, replace at least one of the stored battery capacity value or battery resistance value in the memory of the fuel gauge with at least one of the predicted battery capacity value or the predicted battery resistance value.

[0005] In one example, the present disclosure relates to a method comprising: predicting, by a host device, at least one of a battery capacity or a battery resistance value of a rechargeable battery each time, the rechargeable battery powering the host device; calculating, by the host device, a difference between at least one of the predicted battery capacity value or the predicted battery resistance value and at least one of the battery capacity value or the battery resistance value stored in a memory of a fuel gauge; determining, by the host device, whether the difference exceeds a predetermined maximum error threshold; and replacing, by the host device, at least one of the stored battery capacity value or the battery resistance value in the memory of the fuel gauge with at least one of the predicted battery capacity value or the predicted battery resistance value based on the difference exceeding the predetermined maximum error threshold.

[0006] In one example, the present disclosure relates to a non-transitory storage medium comprising instructions that, when executed by one or more processors, cause the one or more processors to: predict at least one of a battery capacity value or a battery resistance value of a rechargeable battery each time the rechargeable battery powers a host device; calculate a difference between at least one of the predicted battery capacity value or the predicted battery resistance value and a battery capacity value or battery resistance value stored in a memory of a fuel gauge; determine whether the difference exceeds a predetermined maximum error threshold; and based on the difference exceeding the predetermined maximum error threshold, replace at least one of the stored battery capacity value or battery resistance value in the memory of the fuel gauge with at least one of the predicted battery capacity value or the predicted battery resistance value.

[0007] The details of one or more examples of the disclosure are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the disclosure will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 FIG. 4 is a block diagram of a battery management system according to the technology of the present disclosure.

[0009] Figure 2 A curve illustrating the change of battery capacity over time.

[0010] Figure 3 A graph illustrating predicted battery capacity according to the disclosed technique.

[0011] Figure 4 A flow chart illustrating a technique for changing a predicted battery capacity or battery resistance in a battery management system according to the techniques of this disclosure is provided. DETAILED DESCRIPTION

[0012] Various devices may utilize rechargeable batteries as a power source for operating electricity. For example, an implantable medical device (IMD) that provides cardiac rhythm management therapy, monitors one or more physiological parameters of a patient, or provides neurostimulation therapy to a patient may include a rechargeable battery to power the generation of electrical therapy or other functions of the IMD. As another example, a left ventricular assist device (LVAD) may include a rechargeable battery to power the pump and other functions of the LVAD. Rechargeable batteries typically have an integrated circuit, such as a fuel gauge, that monitors the battery and provides information about the battery status to a host device (such as an IMD, laptop, tablet, or cell phone). For example, this information may be the charge level. The fuel gauge may have battery capacity and / or battery resistance values stored in memory, which may have been placed there before the device was deployed. These values can be used by the fuel gauge in a battery management system to present information about the battery status to the user of the host device. However, over time, the charge capacity of a rechargeable battery decreases and its internal resistance increases. Therefore, as the battery ages, the values stored in the fuel gauge's memory may become erroneous. In some situations, such as a lack of deep cycling or continuous cycling without rest, the battery management system may be unable to update the initial battery capacity or battery resistance values. If the stored battery capacity or battery resistance values are incorrect, the information presented to the user regarding the battery status may also be incorrect. For example, the battery management system may overestimate the remaining battery capacity. If the user of the host device relies on this information, they may discover that the battery is depleted before they think it should be.

[0013] Figure 1 FIG. 4 is a block diagram of a battery management system according to the technology of the present disclosure. Figure 1 A host device 8 is depicted, which may be an IMD, another medical device, a laptop, a tablet, a cellular phone, or any other device that utilizes a rechargeable battery to power its operation. Host device 8 includes memory 16, which includes a battery capacity and / or battery resistance prediction algorithm 18. The battery capacity and / or battery resistance prediction algorithm may contain a prediction model 7 of battery capacity and / or battery resistance changes over time. The battery capacity and / or battery resistance prediction algorithm 18 may be implemented in higher-level firmware. In some examples, the battery capacity and / or battery resistance prediction algorithm 18 may be implemented in software rather than firmware.

[0014] The host device 8 also includes one or more processors 4 configured to execute a battery capacity and / or battery resistance prediction algorithm 18. The one or more processors 4 may be implemented as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), field-programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuits. The host device 8 may further include a user interface 2. The user interface 2 may enable a user to provide information to the host device 8 and receive information from the host device 8. For example, the user interface 2 may provide the user of the host device 8 with a visual indication of the battery charge status, as determined by the fuel gauge. The host device 8 may also contain a telemetry interface 5. The telemetry interface 5 may enable the host device 8 to communicate with other devices. For example, the telemetry interface may enable communication using any standardized or proprietary communication protocol, whether wired or wireless. In some examples, when the host device 8 is an IMD, the host device 8 may transmit an indication of the battery charge status to an external device (not shown) via the telemetry interface 5, and the external device may provide the information to the user of the host device 8.

[0015] Figure 1 Also depicted is a battery pack 6. The battery pack 6 includes a fuel gauge 12 and a battery 20. The fuel gauge 12 may be an integrated circuit and may be configured together with the battery 20 in the battery pack 6. The fuel gauge 12 may include a memory 14. The memory 14 may store values associated with the battery 20. For example, the memory 14 may store a battery capacity value 1 and a battery resistance value 3. The battery capacity value 1 and the battery resistance value 3 may be stored in the memory 14 before the battery pack 6 is deployed. The fuel gauge 12 may use the battery capacity value 1 and the battery resistance value 3 to provide information regarding the charge state of the battery 20 to the host device 8 (and, in turn, to the user of the host device 8). The fuel gauge 12 may monitor the battery 20 and estimate the remaining capacity, charge state, or other state parameters of the battery 20 based on the battery capacity, battery resistance, load current, temperature, and other parameters stored in the memory 14. One or more processors 4 executing a battery capacity and / or battery resistance prediction algorithm 18 of the host device 8 may regularly communicate with the fuel gauge 12 to retrieve various battery state parameters and provide information regarding the battery state parameters to the user of the host device 8. For example, one or more processors 4 executing a battery capacity and / or battery resistance prediction algorithm 18 may receive battery state parameters (such as battery capacity 1 or battery resistance 3) from fuel gauge 12 and provide information about the battery state parameters to a user of host device 8 via user interface 2 or telemetry interface 5. Although shown as part of battery management system 10, one or more processors 4 may also be configured to perform other tasks.

[0016] Figure 23 is a graph illustrating the change in battery capacity over time. Battery capacity value 1 represents the battery capacity value stored in memory 14. Curve 32 represents the actual battery capacity of battery 20 as battery 20 ages. As can be seen, the difference between battery capacity value 1 and curve 32 increases over time. Therefore, the older the battery 20, the more likely the host device 8 will receive erroneous information from the fuel gauge 12. Figure 2 Although not depicted, the actual battery resistance also changes as the battery 20 ages. The actual battery resistance increases as the battery 20 ages. If the battery resistance value 3 does not change, the older the battery 20, the more error information the host device 8 receives from the fuel gauge 12 may become.

[0017] Figure 3 Graphs illustrating predicted battery capacity according to the techniques of the present disclosure. Battery capacity value 1 again represents the battery capacity value stored in memory 14. Actual battery capacity is again represented by graph 32. Battery capacity and / or battery resistance algorithm 18 may be used by one or more processors 4 of host device 8 to predict the battery capacity or battery resistance of battery 20.

[0018] exist Figure 3 In the example of , the predicted battery capacity is represented by curve 34. The prediction of battery capacity can be based on a prediction model 7 of battery life, including capacity decay and resistance increase over time. For example, the prediction model 7 can be based on the time and cycle count of the battery. In one example, the prediction model 7 can be created by simulating a battery used for the same or similar purpose (e.g., an IMD). In another example, the prediction model 7 can be based on information collected from batteries previously used for the same or similar purpose. In yet another example, batteries from the same batch as battery 20 can be tested in a laboratory for the same or similar purpose as battery 20, and information about their battery capacity and battery resistance can be periodically provided to the host device via the telemetry interface 5. In this example, the one or more processors 4 can then update the battery capacity and / or battery resistance prediction algorithm 18 based on the test information.

[0019] The one or more processors 4 of the host device 8 may check to determine whether the battery capacity value 1 in the memory 14 is incorrect and may update the battery capacity value 1 to match the predicted capacity. In some examples, the one or more processors 4 of the host device 8 may receive a flag or signal from the fuel gauge 12 indicating that the battery capacity value 1 is incorrect. In other examples, the one or more processors 4 of the host device 8 may determine whether the battery capacity value 1 is incorrect on a periodic basis. The periodic basis may be a fixed time period or may be event-driven, such as checking after each recharge or power-up.

[0020] For example, at time t1, one or more processors 4 may compare the difference between battery capacity value 1 and predicted battery capacity 34. If, at time t1, the difference between battery capacity value 1 and predicted battery capacity 34 is greater than a predetermined maximum error threshold (such as predetermined maximum error threshold 30), one or more processors 4 of host device 8 may write predicted battery capacity 34 at time t1 to memory 14, thereby replacing battery capacity value 1 with battery capacity value 1′. Figure 3 The predetermined maximum error threshold is depicted in a particular manner in FIG. , but this should not be considered limiting. For example, the predetermined maximum error threshold may be greater than or less than the depicted predetermined maximum error threshold. In some examples, the predetermined maximum error threshold may be within 3% of the initial (t0) battery capacity value or a previously updated battery capacity value. In this manner, the information provided by fuel gauge 12 to host device 8 may be more accurate, and any overestimation of the remaining capacity of the battery may be reduced.

[0021] Although Figure 3 The example of relates to a battery capacity value, but the same process can be used for battery resistance value 3 or other battery parameters stored in memory 14. For example, one or more processors 4 of host device 8 may determine that battery resistance value 3 in memory 14 is erroneous and may update the value to match the predicted battery resistance. For example, at time t1, one or more processors 4 may compare the difference between battery resistance value 3 and the predicted battery resistance value. If the difference between battery resistance value 3 and the predicted battery resistance value at time t1 is greater than a predetermined maximum error threshold, one or more processors 4 of host device 8 may write the predicted battery resistance value at time t1 to memory 14, thereby replacing battery resistance value 3 with battery resistance value 3′ (not shown). In some examples, the predetermined maximum error threshold may be within 3% of the initial (t0) battery resistance value or the previously updated battery resistance value.

[0022] Figure 4A flow chart illustrating a technique for changing the predicted battery capacity or battery resistance in a battery management system according to the techniques of the present disclosure. The fuel gauge 12 may have initial parameters, such as a battery capacity value 1 and a battery resistance value 3 stored in a memory 14 (40). The one or more processors 4 may execute the battery capacity and / or battery resistance prediction algorithm 18 at predetermined time intervals. At predetermined time intervals, such as one week, the one or more processors 4 may increment the storage time (42). In some instances, such as when the battery is operated in a high temperature environment, the predetermined time interval may be shorter, causing the one or more processors 4 to execute the battery capacity and / or battery resistance prediction algorithm 18 more frequently. The storage time may be stored in the memory 16 and may be part of the battery capacity and / or battery resistance prediction algorithm 18. Alternatively, the storage time may be stored in the memory 14 of the fuel gauge 12. The one or more processors 4 may then determine whether the battery capacity value 1 and / or battery resistance value 3 stored in the memory 14 has changed due to something other than the battery capacity and / or battery resistance prediction algorithm 18 (44). If the battery capacity value 1 and / or the battery resistance value 3 have changed (the "yes" path of the decision diamond 44), the one or more processors 4 obtain the predicted battery capacity value 1' and / or the predicted battery resistance value 3' as new initial values and update the maximum error threshold value based on the new values (46). For example, if certain conditions are met, such as deep cycling, the battery capacity value 1 and / or the battery resistance value 3 may have been changed by the fuel gauge 12 itself. If the battery capacity value 1 and / or the battery resistance value 3 have not changed (the "no" path of the decision diamond 44), the one or more processors 4 may utilize the battery capacity and / or battery resistance prediction algorithm 18 to predict the battery capacity or battery resistance at that time (e.g., t1) (48). As discussed above, the prediction of the battery capacity or battery resistance may be based on a prediction model 7 of the battery life, including capacity fade and resistance increase over time.

[0023] The one or more processors 4 may calculate the difference between the battery capacity value 1 and / or the battery resistance value 3 stored in the memory 14 and the predicted battery capacity value and / or the predicted battery resistance value (50). The one or more processors 4 may then compare the difference to the maximum error threshold 30 (52). If the difference is greater than the maximum error threshold 30 (the "yes" path of the decision diamond 52), the one or more processors 4 may overwrite the battery capacity value 1 or battery resistance value 3 in the memory 14 of the fuel gauge 12 by writing a predicted battery capacity value 1' or a predicted battery resistance value 3' (not shown) that is higher than the battery capacity value 1 or the battery resistance value 3 (52). The predicted battery capacity value 1' or the predicted battery resistance value 3' (not shown) may then be used as a new initial value at time t0, and the maximum error threshold (46) may be updated based on the new initial value. If the difference does not exceed the maximum error threshold (the "no" path of the decision diamond 52), the one or more processors 4 may return to the decision diamond 44 to determine whether the value stored in the memory 14 has changed.

[0024] Host device 8 may include electronics and other internal components necessary or required to perform the functions associated with the device. In one example, host device 8 includes one or more of the following: processing circuitry, memory, signal generation circuitry, readout circuitry, telemetry circuitry, and a power supply. Generally speaking, the memory of host device 8 may include computer-readable instructions that, when executed by a processor of host device 8, cause the processor to perform various functions attributed to the device herein.

[0025] Host device 8 may include or may be one or more processors or processing circuits, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), field-programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuits. Thus, as used herein, the terms "processor" and "processing circuitry" may refer to any of the foregoing structures or any other structures suitable for implementing the techniques described herein.

[0026] The memory may include any volatile or nonvolatile media such as random access memory (RAM), read only memory (ROM), nonvolatile RAM (NVRAM), electrically erasable programmable ROM (EEPROM), flash memory, etc. The memory may be a storage device or other non-transitory media.

[0027] Various examples have been described in this disclosure. These and other examples are within the scope of the following claims.

Claims

1. A host device, comprising: Memory; and one or more processors coupled to the memory, the one or more processors configured to: predicting at least one of a battery capacity value or a battery resistance value of a rechargeable battery each time, the rechargeable battery powering the host device; calculating a difference between the at least one of the predicted battery capacity value or the predicted battery resistance value and at least one of the battery capacity value or the battery resistance value stored in a memory of the fuel gauge; determining whether the difference exceeds a predetermined maximum error threshold; as well as Based on the difference exceeding the predetermined maximum error threshold, replacing the at least one of the stored battery capacity value or the battery resistance value in the memory of the fuel gauge with the at least one of the predicted battery capacity value or the predicted battery resistance value.

2. The host device of claim 1, wherein the fuel gauge is configured together with the rechargeable battery.

3. The host device of claim 1, wherein the fuel gauge provides an indication of a charge level of the rechargeable battery to the one or more processors. 4 . The host device according to claim 1 , further comprising a prediction model of battery capacity or battery resistance. The host device of claim 4 , wherein the prediction model models changes in battery capacity or battery resistance over time. The host device of claim 4 , wherein the prediction model is implemented in firmware.

7. The host device of claim 6, further comprising an implantable medical device.

8. The host device of claim 4, wherein the one or more processors are further configured to determine whether the at least one of the battery capacity value or the battery resistance value stored in the memory has changed.

9. The host device of claim 8, wherein the one or more processors are further configured to update the maximum error threshold in response to determining that the at least one of the battery capacity value or the battery resistance value stored in the memory has changed.

10. A method for a host device, comprising: predicting, by the host device, at least one of a battery capacity value or a battery resistance value of a rechargeable battery each time, the rechargeable battery powering the host device; calculating, by the host device, a difference between the at least one of the predicted battery capacity value or the predicted battery resistance value and at least one of the battery capacity value or the battery resistance value stored in a memory of the fuel gauge; determining, by the host device, whether the difference exceeds a predetermined maximum error threshold; as well as Based on the difference exceeding the predetermined maximum error threshold, replacing, by the host device, the at least one of the stored battery capacity value or the battery resistance value in the memory of the fuel gauge with the at least one of the predicted battery capacity value or the predicted battery resistance value. The method of claim 10 , wherein the fuel gauge is disposed together with the rechargeable battery.

12. The method of claim 10, further comprising providing, by the fuel gauge, a representation of the charge level of the rechargeable battery to one or more processors of the host device.

13. The method according to any one of claims 10 to 12, wherein the prediction is based on a prediction model of battery capacity or battery resistance. The method of claim 13 , wherein the prediction model models changes in battery capacity or battery resistance over time.

15. A non-transitory storage medium comprising instructions that, when executed by one or more processors, cause the one or more processors to: predicting at least one of a battery capacity value or a battery resistance value of a rechargeable battery each time, the rechargeable battery powering a host device; calculating a difference between the at least one of the predicted battery capacity value or the predicted battery resistance value and the battery capacity value or the battery resistance value stored in a memory of the fuel gauge; determining whether the difference exceeds a predetermined maximum error threshold; as well as Based on the difference exceeding the predetermined maximum error threshold, replacing the at least one of the stored battery capacity value or the battery resistance value in the memory of the fuel gauge with the at least one of the predicted battery capacity value or the predicted battery resistance value.

Images