METHOD OF DETERMINING BATTERY DEGRADATION

Abstract

A method of determining battery degradation retroactively using historical data is disclosed. The method includes the steps of collecting state of charge (SOC) and DC ampere data for a predetermined time period; determining a delta () SOC based on the data collected; creating a set of SOC regimes having a size based on SOC; filtering the SOC data and determining a set of points which indicate a charging or discharging event; and calculating overall Coulombs associated with each charging or discharging event and for each event, producing a timestamp and Coulombs associated with each event.

Claims

1. A method of determining battery degradation retroactively using historical data, comprising the steps of: collecting state of charge (SOC) and DC ampere data for a predetermined time period; determining a delta () SOC based on the data collected; creating a set of SOC regimes having a size based on SOC; filtering the SOC data and determining a set of points which indicate a charging or discharging event; and calculating overall Coulombs associated with each charging or discharging event and for each event, producing a timestamp and Coulombs associated with each event.

2. The method according to claim 1, wherein the step of collecting includes the step of collecting raw, high temporal-resolution SOC and DC ampere data.

3. The method according to claim 2, wherein the high temporal-resolution SOC and DC ampere data is collected as a time-series data of one second or less.

4. The method according to claim 2, wherein the predetermined time is about one-year.

5. The method according to claim 1, wherein the step of determining further includes the step of selecting a SOC greater than SOC.

6. The method according to claim 1, wherein the step of filtering further includes the step of filtering the raw data to only include data where min_SOC.sub.regime n<SOC<max_SOC.sub.regime n.

7. The method according to claim 1, further including the step of calculating Coulombs lost per day.

8. A method of determining battery degradation retroactively using historical data, comprising the steps of: collecting raw, high temporal-resolution state of charge (SOC) and DC ampere data of an entire energy storage system for a predetermined time period; determining a delta () SOC where SOC is greater than SOC; creating a set of SOC regimes having a size based on SOC; for each SOC regime, filtering the raw SOC data to only include data where min_SOC.sub.regime n<SOC<max_SOC.sub.regime n; after filtering, determining a set of points which indicate a charging or discharging event; and calculating overall Coulombs associated with each charging or discharging event and for each event, producing a timestamp and Coulombs associated with each event.

9. The method according to claim 8, wherein the indication of a charging or discharging event is provided by a set of points without missing data.

10. The method according to claim 8, further including the step of calculating Coulombs lost per day.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The invention may be best understood by reference to the following description taken in conjunction with the accompanying drawing figures, in which:

[0010] FIG. 1 shows a method according to an embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION

[0011] Referring to the drawings wherein identical reference numerals denote the same elements throughout the various views, FIG. 1 illustrates a method of determining battery degradation and is shown generally at reference numeral 10. In general, the method 10 allows for independent verification of storage performance (as opposed to vendor self-reporting) which then allows for better identification of operating costs, best operation and maintenance practices, and enforcement of vendor warranties, as well as, improved specifications and test procedures for future systems.

[0012] The method 10 uses a retroactive analysis methodology that utilizes Battery Management System (BMS)-reported state-of-charge (SOC) and DC ampere measurements from historical battery data, over a period of time (months to years), to derive (1) information about the BMS's SOC estimation algorithm and (2) a simple degradation metric that may indicate the battery system's State-of-Health (SOH). This methodology looks at the number of coulombs (i.e. the integral of the current) that are needed to change the BMS-reported SOC value by the BMS's smallest discernable increment (i.e. the BMS-reported SOC resolution). The change over time in the coulombs required to change the BMS-reported SOC can provide information about a loss of battery capacity (i.e. degradation) and can be used to identify inconsistencies and changes in how the BMS calculates the battery's SOC. This methodology benefits from long-duration (months to years) and high temporal-resolution (sub-five minutes) historical data. It is applicable at any point within an energy storage system where State-of-Charge and DC amperes measurements are reported and is agnostic to the electricity-in, electricity-out energy storage technology.

[0013] More particularly, as shown in FIG. 1 (Block 11), the method 10 starts with approximately one-year of raw, high temporal-resolution (ideally 1 second or better) SOC (%) and DC current (amperes) timeseries data for an entire energy storage system, rack, module, or cell. As illustrated (Block 12), based on the precision of the SOC, choose a SOC value where SOC>SOC precision, e.g. if the SOC is reported in 0.5% increments, then choose a SOC>0.5%; where

[00001]$\frac{\mathrm{SOC}}{0.5\%}=\mathrm{integer}.$

Next (Block 13), create a set of SOC regimes where each regime has a size SOC, e.g. if we choose SOC=1%, there would be 199 regimes: 0%-1.0%, 0.5%-1.5%, 1.0%-2.0%, . . . , 99.0%-100%. For each SOC regime (Block 14), filter the raw data to only include data where min_SOC.sub.regime n<SOC<max_SOC.sub.regime n. Continuing the above example, for the 101.sup.st regime, the data would be filtered such that only data where 50.0%<SOC<51.0% would move onto the next step.

[0014] After filtering (Block 16), consider each discrete set of points without missing data to be a distinct charging or discharging event (charge vs discharge is indicated by the sign of the average current during the event). Integrate the current (amps=coulombs/second) for each event to calculate the overall coulombs associated with each charging or discharging event. Other filtering criteria can be added on top of the no missing data criterium. For instance, a threshold for average current might be used to eliminate events where the drop in SOC was likely caused by self-discharge rather than active discharge. For each charge or discharge event (Block 17), a timestamp (the first datetime in the event set) and the coulombs associated with this event are produced. Apply a regression model to this data to calculate a coulombs lost/day metric for the 101.sup.st SOC regime.

[0015] Alternatively, the method 10 for retroactively estimating battery energy storage SOH may be modified so that no part of the method relies on a hidden BMS calculation like State-of-Charge. This modified method is meant to be applied to a fielded battery energy storage system that does not regularly undergo capacity tests. By aggregating the effect of charge entering and leaving the battery system on the battery system's open circuit voltage across a large number of charge-discharge cycles of different depths-of-discharge and at different voltage levels, a single number for the remaining charge capacity of the battery storage system can be generated, which is a key indicator of SOH. Additionally, the number of coulombs of charge that can be discharged between any given starting open circuit voltage and the system's minimum open circuit voltage can be established, which can be used to retroactively estimate the battery system's SOC. Combining this with a well-trained equivalent circuit model can yield a reliable, transparent estimate of the expected amount of energy yield under different discharge conditions from different starting SOCs. The same process can be applied to the charging half-cycles to estimate the amount of grid energy required to charge the battery system from an arbitrary starting SOC to an arbitrary ending SOC.

[0016] The foregoing has described a method of determining battery degradation and SOC retroactively using historical data. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive.

[0017] Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

[0018] The invention is not restricted to the details of the foregoing embodiment(s). The invention extends any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.