Method of evaluating battery state of charge

Abstract

A method of evaluating battery state of charge includes steps of: setting a candidate state of charge; measuring practical released voltages of a currently used battery respectively at a plurality of rebound time points after the currently used battery is charged or discharged; obtaining evaluated released voltages of the currently used battery having the candidate state of charge; calculating differences between the practical released voltages and the evaluated released voltages; summing up the differences to obtain a total estimation error value; determining whether or not the total estimation error value is larger than a threshold, if not, determining that a current state of charge of the currently used battery is equal to or approximately equal to the candidate state of charge, if yes, changing the candidate state of charge in the first one of the above-mentioned steps and performing the subsequent steps based on the changed candidate state of charge.

Description

Technical Field

[0001] This invention relates to batteries, and more particularly to a method for assessing the remaining percentage of battery charge. Prior Technology

[0002] With advancements in technology and a growing awareness of environmental protection, electric vehicles have become one of the mainstream green transportation options in various countries. Since the primary power source for electric vehicles is the battery, battery management has always been a crucial issue for electric vehicle designers.

[0003] In battery management, accurate State of Charge (SOC) estimation allows users to know when to charge the battery and prevents electric vehicles from losing power and stopping, or even posing a danger. However, after prolonged use, errors accumulated in the battery charge estimation process render the SOC estimates from traditional battery management methods inaccurate and unreliable, requiring correction using steady-state voltage. However, measuring the steady-state voltage requires the battery to be left idle for an extended period after charging or discharging, making traditional battery management methods time-consuming to obtain an accurate SOC percentage. Therefore, accurately estimating the remaining battery charge percentage well before the long idle period ends after charging or discharging is a significant challenge for those skilled in the art. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for evaluating the remaining battery capacity percentage, comprising the following steps: (a) selecting one from a plurality of specified remaining capacity percentages as a candidate remaining capacity percentage; (b) measuring multiple release voltages at multiple rebound time points of a practical battery during a resting period after charging or discharging, as multiple practical release voltages; (c) estimating multiple voltages at multiple rebound time points of the practical battery with the candidate remaining capacity percentage after charging or discharging, as multiple estimated release voltages; and (d) calculating the multiple practical release voltages relative to the given voltages. Multiple differences between multiple estimated release voltages, or multiple squares of the multiple differences, are used as multiple estimated error values; and (e) it is determined whether each estimated error value is greater than an error threshold or whether the total estimated error value obtained by summing the multiple estimated error values ​​at multiple rebound time points is greater than a total error threshold. If not, an evaluation error value between the candidate remaining charge percentage and a current remaining charge percentage of the practical battery is evaluated to be less than an evaluation error threshold. If so, return to step (a) but select another of the multiple specified remaining charge percentages as the candidate remaining charge percentage.

[0005] As described above, the present invention provides a method for assessing the remaining battery charge percentage. After the battery is charged or discharged, it does not require waiting for a long period of complete rest time until the battery's steady-state voltage is obtained, as is the case with traditional battery management methods. Instead, it can quickly and accurately estimate the remaining battery charge percentage in advance during a short period of rest time, that is, before the end of the battery's complete rest time. Therefore, it can manage the battery's state more instantly. Simple Explanation of the Diagram

[0006] Figure 1 is a flowchart of the steps of the battery remaining power percentage assessment method according to the first embodiment of the present invention.

[0007] Figure 2 is a schematic diagram of the line segments of the sampled curves obtained by the battery remaining power percentage assessment method of the second embodiment of the present invention.

[0008] Figure 3 is a schematic diagram of the curve of the change of battery release voltage with resting time as measured and estimated by the battery remaining power percentage assessment method of the second embodiment of the present invention.

[0009] Figure 4 is a schematic diagram of the curve of the total estimated error value and the remaining power percentage after multiple tracking of the battery remaining power percentage assessment method of the second embodiment of the present invention.

[0010] Figure 5 is a schematic diagram of the line segments of the sampled curves obtained by the battery remaining power percentage assessment method of the third embodiment of the present invention.

[0011] Figure 6 is a schematic diagram of the curves showing the change of battery release voltage with resting time as measured and estimated by the battery remaining power percentage assessment method of the third embodiment of the present invention.

[0012] Figure 7 is a schematic diagram of the curve of the total estimated error value and the remaining power percentage after multiple tracking of the battery remaining power percentage assessment method of the third embodiment of the present invention.

[0013] Figure 8 is a schematic diagram of the line segments of the sampled curves obtained by the battery remaining power percentage assessment method according to the fourth embodiment of the present invention.

[0014] Figure 9 is a schematic diagram of the curves showing the change of battery release voltage with resting time as measured and estimated by the battery remaining power percentage assessment method of the fourth embodiment of the present invention.

[0015] Figure 10 is a schematic diagram of the curve of the total estimated error value and the remaining power percentage after multiple tracking of the battery remaining power percentage assessment method according to the fourth embodiment of the present invention.

[0016] Figure 11 is a block diagram of the battery remaining power percentage assessment system according to the first to fourth embodiments of the present invention. Implementation

[0017] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the concept of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions, as stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention. In addition, the term "or" used herein should be interpreted to include, depending on the actual situation, any combination of any one or more of the associated listed items.

[0018] Please refer to Figures 1 and 11, where Figure 1 is a flowchart of the steps of the battery remaining power percentage assessment method according to the first embodiment of the present invention, and Figure 11 is a block diagram of the battery remaining power percentage assessment system according to the first to fourth embodiments of the present invention.

[0019] In the first embodiment, the battery remaining power percentage assessment method of the present invention includes steps S101 to S108 as shown in FIG1. ​​These steps can be executed by multiple hardware components or circuits of the battery remaining power percentage assessment system of the present invention, including processors, controllers, sensors, and other hardware devices. For example, but not limited to, they can be executed by all or some of the remaining power percentage candidate module 100, battery voltage measurement module 200, battery data prediction module 300, prediction error calculation module 400, remaining power percentage judgment module 500, and tracking curve construction module 600 as shown in FIG11. In practice, the tracking curve construction module 600 can be omitted, and its execution can be performed by other modules of the battery remaining power percentage assessment system of the present invention or other external devices; the present invention is not limited thereto.

[0020] In step S101, the remaining power percentage candidate module 100 selects one of a plurality of specified remaining power percentages within a specified remaining power percentage range (e.g., 0% to 100%) as a candidate remaining power percentage.

[0021] In step S102, the battery voltage measurement module 200 measures multiple release voltages at multiple rebound time points of the practical battery during a resting period after charging or discharging, and these are taken as multiple practical release voltages. To facilitate the distinction between different batteries, in this paper, the battery used for testing, estimation, and model building is referred to as the test battery, while the battery actually used is referred to as the practical battery.

[0022] In step S103, the battery data prediction module 300 predicts multiple voltages at multiple rebound time points of the practical battery with candidate remaining charge percentages after charging or discharging, as multiple predicted release voltages.

[0023] For example, the battery remaining capacity percentage assessment method of the present invention may further include the following steps that can be executed by the battery data prediction module 300: based on the multiple release voltages and steady-state voltages at multiple rebound time points when the test battery has multiple specified remaining capacity percentages after charging or discharging, respectively, multiple model training parameter sets generated based on the multiple specified remaining capacity percentages; and training a battery voltage prediction model, respectively, based on the multiple model training parameter sets generated based on the multiple specified remaining capacity percentages, predicting multiple release voltages at multiple rebound time points of the practical battery during a rest period after charging or discharging to a candidate remaining capacity percentage, as multiple predicted release voltages.

[0024] For example, the battery remaining capacity percentage assessment method of the present invention may further include the following steps, which can be executed by the battery data prediction module 300, of training a battery voltage prediction model: obtaining a second-order training model equation as a battery voltage prediction model; and substituting a resting time between a charging end time point or a discharging end time point of the practical battery and each rebound time point, the release voltage of the practical battery at each rebound time point, and the steady-state voltage of the practical battery reaching steady state into the following second-order training model equation: , in Representing each release voltage, This represents the resting time between the end of battery charging or discharging and each rebound time point. Represents steady-state voltage. , , , This represents multiple model training parameters. The model training parameter set generated above based on each specified remaining battery percentage can contain multiple model parameters V1, V2, ... .

[0025] The above example uses a second-order model, but in practice, more orders can be used. The order can depend on the battery's constituent materials. The following battery data prediction module 300 obtains a multi-order training model equation as the battery voltage prediction model. The following multi-order training model equation is substituted into the following equation: a resting time between a charging end time point or a discharging end time point and each rebound time point of the practical battery, the release voltage of the practical battery at each rebound time point, and the steady-state voltage of the practical battery reaching steady state. … , in Represents the release voltage. This represents the resting time between the end of the battery charging or discharging process and each rebound time point. Represents steady-state voltage. ~ as well as ~ This represents multiple model training parameters. The model training parameter set generated above based on each specified remaining battery percentage can contain multiple model training parameters. ~ as well as ~ .

[0026] In step S104, the prediction error calculation module 400 calculates a difference between a practical release voltage and a predicted release voltage at the same rebound time point. In the second to fourth embodiments shown in Figures 4, 7 and 10, the square of the difference is used as a prediction error value. In this way, multiple differences or squares of multiple differences between multiple practical release voltages at multiple rebound time points and multiple predicted release voltages are calculated as multiple prediction error values.

[0027] In step S105, the prediction error calculation module 400 sums up the multiple prediction error values ​​at multiple rebound time points to obtain a total prediction error value.

[0028] In step S106, the remaining power percentage determination module 500 determines whether a total estimated error value is greater than a total error threshold value.

[0029] If the total estimated error value is greater than a total error threshold value, step S107 is executed. In step S107, the remaining battery percentage determination module 500 instructs the remaining battery percentage candidate module 100 to select another among multiple specified remaining battery percentages within a specified remaining battery percentage range (e.g., 0%~100%) as a candidate remaining battery percentage. Then, based on the other among multiple specified remaining battery percentages selected next time as a candidate remaining battery percentage, steps S103~S106 are executed sequentially.

[0030] Conversely, if the remaining battery percentage determination module 500 determines that the total estimated error value is not greater than a total error threshold value, step S108 is executed. In step S108, the remaining battery percentage determination module 500 determines that the candidate remaining battery percentage selected by the candidate remaining battery percentage module 100 is equal to or close to the current remaining battery percentage (State of Charge, SOC) of the actual battery, that is, the error value (as an evaluation error value) between the currently selected candidate remaining battery percentage and the current remaining battery percentage of the actual battery is less than an evaluation error threshold value. Since the battery remaining battery percentage evaluation method of the present invention can accurately evaluate the candidate remaining battery percentage that is equal to or close to the current remaining battery percentage of the actual battery, the candidate remaining battery percentage selected in step S108 can be used as the current remaining battery percentage of the actual battery.

[0031] In this embodiment, an example is to compare a total estimated error value and a total error threshold value at multiple rebound time points, but the present invention is not limited thereto. In practice, step S105 can be omitted, and step S106 can be replaced by the remaining battery percentage judgment module 500 judging whether the estimated error value of each or any one of the multiple estimated error values ​​at multiple rebound time points is greater than an error threshold value. In this case, if the remaining battery percentage judgment module 500 judges that the estimated error value of each or any one of the multiple estimated error values ​​at multiple rebound time points is greater than an error threshold value, step S107 is executed. Conversely, if the remaining battery percentage judgment module 500 judges that the estimated error value of each or any one of the multiple estimated error values ​​at multiple rebound time points is not greater than an error threshold value, step S108 is executed.

[0032] Please refer to Figures 2 to 4 and Figure 11, where Figure 2 is a schematic diagram of the line segments of the sampled curve of the battery remaining power percentage assessment method of the second embodiment of the present invention, Figure 3 is a schematic diagram of the curve of the change of battery release voltage with resting time measured and estimated by the battery remaining power percentage assessment method of the second embodiment of the present invention, and Figure 4 is a schematic diagram of the curve of the total estimated error value and the remaining power percentage tracked multiple times by the battery remaining power percentage assessment method of the second embodiment of the present invention.

[0033] It should be understood that after a battery is charged or discharged, it is left to rest for a period of time as a resting period. During this resting period, the battery charging / discharging is stopped. During this resting period, the battery voltage will rebound. After a period of time, the battery voltage will reach a steady-state voltage and remain at this steady-state voltage.

[0034] For example, the battery voltage rebound curve in Figure 3 is a battery voltage rebound curve established based on multiple discharge voltages that gradually rebound / increase over time during a rest period after discharge of a practical battery.

[0035] It should be understood that while users can measure the battery voltage, they may find it difficult to accurately determine the remaining state of charge (SOC). However, the battery remaining state of charge assessment method of the present invention only requires sampling a small amount of battery data to determine the remaining state of charge, as detailed below.

[0036] In the battery remaining capacity percentage assessment method of this invention, multiple voltages from one of the multiple line segments contained in a battery voltage rebound curve can be sampled. For example, but not limited to, multiple release voltages and respective resting times represented by a line segment in the curve of Figure 2 within a resting time period, for subsequent analysis to determine the remaining capacity percentage of the practical battery. It is worth noting that this invention does not require waiting a long time for the battery to reach a steady state to obtain a steady-state voltage. Instead, it only needs to sample data before the practical battery reaches a steady state to predict the remaining capacity percentage of the practical battery. This is done by gradually tracking the remaining capacity percentage until it is closest to the actual remaining capacity percentage of the practical battery.

[0037] In the first tracking, the remaining charge percentage candidate module 100 randomly selects one of several specified remaining charge percentages within a specified remaining charge percentage range (e.g., 0%~100%), such as "90%" as shown in Figure 4, as a candidate remaining charge percentage. That is, the initial assumption is that the remaining charge percentage (SOC) of the actual battery is, for example, "90%". The prediction error calculation module 400 uses the model training parameter set generated based on the candidate remaining charge percentage, for example, "90%", in a battery voltage prediction model to predict multiple release voltages of the actual battery at multiple rebound time points as multiple predicted release voltages. The tracking curve construction module 600 can build a battery rebound voltage curve, where the horizontal axis represents multiple resting times and the vertical axis represents multiple actual release voltages and multiple predicted release voltages. The tracking curve construction module 600 constructs the battery rebound voltage curve of the first tracking, as shown in Figure 3, based on these multiple predicted release voltages in the battery rebound voltage curve.

[0038] When the total estimated error value of multiple release voltages and multiple estimated voltages of the practical battery represented by the rebound voltage curve of a candidate remaining power percentage "90%" exceeds a total error threshold value (as shown in step S106 in Figure 1), the remaining power percentage judgment module 500 determines that the "90%" selection / assumption is incorrect. Therefore, a second tracking is performed.

[0039] In the battery remaining power percentage assessment method of the present invention, the tracking curve construction module 600 may also perform the following steps: establish a current remaining power percentage tracking chart as shown in FIG4, wherein multiple total estimated error values ​​are marked on one of the horizontal axis and the vertical axis of the battery remaining power percentage tracking chart and multiple remaining power percentages are marked on the other axis; based on a candidate remaining power percentage selected for the first time and its total estimated error value, mark or construct a tracking pattern on the current remaining power percentage tracking chart, such as the dots shown in FIG4.

[0040] In the second tracking, the remaining battery percentage candidate module 100 calculates a candidate remaining battery percentage, such as "70%" as shown in Figure 4, based on the total prediction error value calculated in the first tracking. The prediction error calculation module 400 uses the model training parameter set generated based on the candidate remaining battery percentage, such as "70%", in a battery voltage prediction model to predict multiple release voltages of the actual battery at multiple rebound time points as multiple predicted release voltages. The tracking curve construction module 600 then constructs the battery rebound voltage curve of the second tracking, as shown in Figure 3, in a battery rebound voltage curve graph.

[0041] When the total estimated error value of multiple release voltages and multiple estimated voltages of the practical battery represented by the rebound voltage curve of a candidate remaining power percentage of "70%" exceeds a total error threshold value (as shown in steps S102~S106 in Figure 1), the remaining power percentage judgment module 500 determines that the "70%" selection / assumption is incorrect. Therefore, a third trace is performed.

[0042] In the battery remaining power percentage assessment method of the present invention, the tracking curve construction module 600 may also perform the following steps: based on the candidate remaining power percentage selected each time after the first selection and its total estimated error value, move the position of a tracking pattern on a battery remaining power percentage chart or mark the next / another tracking pattern on the battery remaining power percentage chart; mark the moving direction of a tracking pattern or mark the moving direction of the next tracking pattern compared to the previous tracking pattern, for example, as shown in FIG4, an arrow points to the moving direction of a tracking pattern marked with a dot.

[0043] In the battery remaining power percentage assessment method of the present invention, the following tracking steps may be included: the prediction error calculation module 400 determines whether a total prediction error value calculated based on another of a plurality of specified remaining power percentages is less than a total prediction error value calculated based on one of the plurality of specified remaining power percentages.

[0044] If the total estimated error value calculated based on another of the multiple specified remaining battery percentages is less than the total estimated error value calculated based on one of the multiple specified remaining battery percentages, the estimated error calculation module 400 determines that the tracking direction is correct, and in the next tracking (as shown in step S101 in Figure 1), selects another remaining battery percentage that is closer to the other than the other from the multiple specified remaining battery percentages within a specified remaining battery percentage range (e.g., 0%~100%) as a candidate remaining battery percentage.

[0045] Conversely, if the total estimated error value calculated based on another of the multiple specified remaining battery percentages is not less than the total estimated error value calculated based on one of the multiple specified remaining battery percentages, the estimated error calculation module 400 determines that the tracking direction is wrong, and in the next tracking (as shown in step S101 in Figure 1), selects another specified remaining battery percentage that is closer to one of the multiple specified remaining battery percentages within a specified remaining battery percentage range (e.g., 0%~100%) as a candidate remaining battery percentage.

[0046] For example, as shown in Figure 4, during the second to sixth tracking iterations, the remaining battery percentage is assumed to be "70%" → (decreasing) "30%" → (decreasing) "20%" → (increasing) "25%" → (decreasing) "22.5%". The sixth tracking iteration shows a remaining battery percentage of 22.5%, which is close to the actual remaining battery percentage of "22.44%". If tracking continues for several more iterations, the actual remaining battery percentage can be accurately tracked.

[0047] Please refer to Figures 5 to 7 and Figure 11, where Figure 5 is a schematic diagram of the line segments of the sampled curve of the battery remaining power percentage assessment method of the third embodiment of the present invention, Figure 6 is a schematic diagram of the curve of the change of battery release voltage with resting time measured and estimated by the battery remaining power percentage assessment method of the third embodiment of the present invention, and Figure 7 is a schematic diagram of the curve of the total estimated error value and the remaining power percentage tracked multiple times by the battery remaining power percentage assessment method of the third embodiment of the present invention.

[0048] In the battery remaining power percentage assessment method of the present invention, multiple voltages of one of the multiple line segments contained in a battery voltage rebound curve can be sampled, for example, but not limited to, multiple release voltages and respective resting times represented by a line segment in the curve of Figure 5 within a resting time, for subsequent analysis to determine the remaining power percentage of the practical battery.

[0049] In the first tracking, the remaining battery percentage candidate module 100 randomly selects one of several specified remaining battery percentages within a specified remaining battery percentage range (e.g., 0%~100%), such as "94.95%" as shown in Figure 7, as a candidate remaining battery percentage. That is, the initial assumption is that the remaining battery percentage of the actual battery is, for example, "94.95%". The prediction error calculation module 400 uses the model training parameter set generated based on the candidate remaining battery percentage, for example, "94.95%", in a battery voltage prediction model to predict multiple release voltages of the actual battery at multiple rebound time points as multiple predicted release voltages. The tracking curve construction module 600 can build a battery rebound voltage curve, where the horizontal axis represents multiple resting times and the vertical axis represents multiple actual release voltages and multiple predicted release voltages. The tracking curve construction module 600 constructs the battery rebound voltage curve of the first tracking, as shown in Figure 6, based on these multiple predicted release voltages in the battery rebound voltage curve.

[0050] When the total estimated error value of multiple release voltages and multiple estimated voltages of the practical battery represented by the rebound voltage curve of a candidate remaining charge percentage of "94.95%" exceeds a total error threshold value (as shown in step S106 in Figure 1), the remaining charge percentage judgment module 500 determines that the "94.95%" selection / assumption is incorrect. Therefore, a second tracking is performed.

[0051] In the battery remaining power percentage assessment method of the present invention, the tracking curve construction module 600 may also perform the following steps: establish a current remaining power percentage tracking chart as shown in FIG7, wherein multiple total estimated error values ​​are marked on one of the horizontal axis and the vertical axis of the battery remaining power percentage tracking chart and multiple remaining power percentages are marked on the other axis; based on a candidate remaining power percentage selected for the first time and its total estimated error value, mark or construct a tracking pattern on the current remaining power percentage tracking chart, such as the dots shown in FIG7.

[0052] In the second tracking, the remaining battery percentage candidate module 100 randomly selects one of several specified remaining battery percentages within a specified remaining battery percentage range (e.g., 0%~100%), such as "92.71%" as shown in Figure 7, as a candidate remaining battery percentage. That is, the second assumption is that the remaining battery percentage of the actual battery is "92.71%". The prediction error calculation module 400 uses the model training parameter set generated based on a candidate remaining battery percentage, such as "92.71%", in a battery voltage prediction model to predict multiple release voltages of the actual battery at multiple rebound time points as multiple predicted release voltages. The tracking curve construction module 600 constructs the battery rebound voltage curve of the second tracking as shown in Figure 6 in a battery rebound voltage curve graph.

[0053] When the total estimated error value of multiple release voltages and multiple estimated voltages of a practical battery represented by a battery rebound voltage curve representing a candidate remaining charge percentage of "92.71%" exceeds a total error threshold value (as shown in steps S102~S106 in Figure 1), the remaining charge percentage judgment module 500 determines that the "92.71%" selection / assumption is incorrect. Therefore, a third tracking is performed.

[0054] In the battery remaining power percentage assessment method of the present invention, the tracking curve construction module 600 may also perform the following steps: based on the candidate remaining power percentage selected each time after the first selection and its total estimated error value, move the position of a tracking pattern on a battery remaining power percentage chart or mark the next / another tracking pattern on the battery remaining power percentage chart; mark the moving direction of a tracking pattern or mark the moving direction of the next tracking pattern relative to the previous tracking pattern, for example, as shown in FIG7, an arrow points to the moving direction of a tracking pattern marked with a dot.

[0055] As shown in Figure 7, during the second to sixth tracking iterations, the remaining battery percentage is assumed to be: 92.71% → 89.86% (decreasing) → 22.29% (decreasing) → 34.35% (increasing) → 35.44% (increasing). The sixth tracking iteration shows a remaining battery percentage of 35.44%, which is close to the actual remaining battery percentage of 35.50%. With further tracking iterations, the actual remaining battery percentage can be accurately tracked.

[0056] Please refer to Figures 8 to 11, where Figure 8 is a schematic diagram of the line segment of the sampled curve of the battery remaining power percentage assessment method of the fourth embodiment of the present invention, Figure 9 is a schematic diagram of the curve of the change of battery release voltage with resting time measured and estimated by the battery remaining power percentage assessment method of the fourth embodiment of the present invention, and Figure 10 is a schematic diagram of the curve of the total estimated error value and the remaining power percentage tracked multiple times by the battery remaining power percentage assessment method of the fourth embodiment of the present invention.

[0057] In the battery remaining power percentage assessment method of the present invention, multiple voltages of one of the multiple line segments contained in a battery voltage rebound curve can be sampled, for example, but not limited to, multiple release voltages and respective resting times represented by a line segment in the curve of Figure 8 within a resting time, for subsequent analysis to determine the remaining power percentage of the practical battery.

[0058] In the first tracking, the remaining battery percentage candidate module 100 randomly selects one of several specified remaining battery percentages within a specified remaining battery percentage range (e.g., 0%~100%), such as "90%" as shown in Figure 10, as a candidate remaining battery percentage. That is, the remaining battery percentage of the actual battery is initially assumed to be, for example, "90%". The prediction error calculation module 400 uses the model training parameter set generated based on the candidate remaining battery percentage, for example, "90%", in a battery voltage prediction model to predict multiple release voltages of the actual battery at multiple rebound time points as multiple predicted release voltages. The tracking curve construction module 600 can build a battery rebound voltage curve, where the horizontal axis represents multiple resting times and the vertical axis represents multiple actual release voltages and multiple predicted release voltages. The tracking curve construction module 600 constructs the battery rebound voltage curve of the first tracking, as shown in Figure 9, based on these multiple predicted release voltages in the battery rebound voltage curve.

[0059] When the total estimated error value of multiple release voltages and multiple estimated voltages of the practical battery represented by the rebound voltage curve of a candidate remaining power percentage "90%" exceeds a total error threshold value (as shown in step S106 in Figure 1), the remaining power percentage judgment module 500 determines that the "90%" selection / assumption is incorrect. Therefore, a second tracking is performed.

[0060] In the battery remaining power percentage assessment method of the present invention, the following steps may also be performed by the tracking curve construction module 600: establishing a current remaining power percentage tracking chart as shown in FIG10, wherein multiple total estimated error values ​​are marked on one of the horizontal axis and the vertical axis of the battery remaining power percentage tracking chart and multiple remaining power percentages are marked on the other axis; based on a candidate remaining power percentage selected for the first time and its total estimated error value, marking or constructing a tracking pattern on the current remaining power percentage tracking chart, such as the dots shown in FIG10.

[0061] In the second tracking, the remaining battery percentage candidate module 100 randomly selects one of multiple specified remaining battery percentages within a specified remaining battery percentage range (e.g., 0%~100%), such as "70%" as shown in Figure 10, as a candidate remaining battery percentage. That is, the second assumption is that the remaining battery percentage of the actual battery is "70%". The prediction error calculation module 400 predicts multiple release voltages of the actual battery at multiple rebound time points based on the model training parameter set generated in a battery voltage prediction model based on a candidate remaining battery percentage, such as "70%". The tracking curve construction module 600 constructs the battery rebound voltage curve of the second tracking as shown in Figure 9 in a battery rebound voltage curve graph.

[0062] When the total estimated error value of multiple release voltages and multiple estimated voltages of the practical battery represented by the rebound voltage curve of a candidate remaining power percentage of "70%" exceeds a total error threshold value (as shown in steps S102~S106 in Figure 1), the remaining power percentage judgment module 500 determines that the "70%" selection / assumption is incorrect. Therefore, a third trace is performed.

[0063] In the battery remaining power percentage assessment method of the present invention, the tracking curve construction module 600 may also perform the following steps: based on the candidate remaining power percentage selected each time after the first selection and its total estimated error value, move the position of a tracking pattern on a battery remaining power percentage chart or mark the next / another tracking pattern on the battery remaining power percentage chart; mark the moving direction of a tracking pattern or mark the moving direction of the next tracking pattern relative to the previous tracking pattern, for example, as shown in FIG10, an arrow points to the moving direction of a tracking pattern marked with a dot.

[0064] As shown in Figure 10, during the second to seventh tracking iterations, the remaining battery percentage is assumed to be: 70% → 30% (decreasing) → 10% (decreasing) → 15% (increasing) → 17.5% (increasing) → 16.23% (decreasing). The seventh tracking iteration shows a remaining battery percentage of 16.23%, which is close to the actual remaining battery percentage of 15.27%. With further tracking iterations, the actual remaining battery percentage can be accurately tracked.

[0065] It is worth noting that the second to fourth embodiments of the present invention use batteries with different material compositions and different characteristics for testing. Based on the above, it is clear that the battery remaining power percentage assessment method of the present invention is applicable to various batteries with different material compositions and different characteristics, and can quickly and accurately predict the remaining power percentage of various batteries in advance.

[0066] In summary, this invention provides a method for assessing the remaining battery charge percentage. Unlike traditional battery management methods that require waiting a long period of complete resting time after charging or discharging to obtain the battery's steady-state voltage, this method can quickly and accurately estimate the remaining battery charge percentage during a short resting period, before the end of the battery's complete resting time. This allows for more timely control of the battery's status.

[0067] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of the patent application of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the scope of the patent application of the present invention.

[0068] S101~S108: Steps 100: Remaining battery percentage candidate modules 200: Battery Voltage Measurement Module 300: Battery Data Prediction Module 400: Prediction Error Calculation Module 500: Remaining battery percentage determination module 600: Tracing Curve Construction Module

Claims

1. A method for assessing the remaining charge percentage of a battery, comprising the following steps: (a) selecting one from a plurality of specified remaining charge percentages as a candidate remaining charge percentage; (b) measuring a plurality of release voltages at a plurality of rebound time points of a practical battery during a resting period after charging or discharging, as a plurality of practical release voltages; (c) estimating a plurality of voltages at the plurality of rebound time points of the practical battery having the candidate remaining charge percentage after charging or discharging, as a plurality of estimated release voltages; (d) calculating a plurality of differences or a plurality of squares of the plurality of differences between the plurality of practical release voltages and the plurality of estimated release voltages, as a plurality of estimation error values; and (e) Determine whether each of the estimated error values ​​is greater than an error threshold or whether the total estimated error value obtained by summing the estimated error values ​​of the multiple rebound time points is greater than a total error threshold. If not, assess whether the evaluation error value between the candidate remaining power percentage and the current remaining power percentage of the practical battery is less than an evaluation error threshold. If so, return to step (a) but select another one of the multiple specified remaining power percentages as the candidate remaining power percentage.

2. The battery remaining power percentage assessment method as described in claim 1 further includes the following steps: training a battery voltage prediction model, and predicting multiple predicted release voltages based on multiple model training parameter sets generated based on multiple specified remaining power percentages.

3. The battery remaining power percentage assessment method as described in claim 2 further includes the following steps: determining whether the total estimated error value calculated based on another of the plurality of specified remaining power percentages is less than the total estimated error value calculated based on one of the plurality of specified remaining power percentages; if so, in the next execution of step (a), selecting another one from the plurality of specified remaining power percentages that is closer to the other than the other as the candidate remaining power percentage; if not, in the next execution of step (a), selecting another one from the plurality of specified remaining power percentages that is closer to the other than the other as the candidate remaining power percentage.

4. The battery remaining capacity percentage assessment method as described in claim 2 further includes the following steps: obtaining a second-order training model equation as the battery voltage prediction model; and substituting a resting time between a charging end time point or a discharging end time point of the practical battery and each of the rebound time points, the release voltage of the practical battery at each of the rebound time points, and the steady-state voltage of the practical battery reaching a steady state into the following second-order training model equation: , where represents each of the release voltages, represents the resting time, represents the steady-state voltage, and , , , represent multiple model training parameters; wherein the model training parameter group generated based on each specified remaining capacity percentage includes multiple model training parameters.

5. The battery remaining capacity percentage assessment method as described in claim 2 further includes the following steps for training the battery voltage prediction model: obtaining a multi-order training model equation as the battery voltage prediction model; and substituting a resting time between a charging end time point or a discharging end time point of the practical battery and each of the rebound time points, the release voltage of the practical battery at each of the rebound time points, and the steady-state voltage of the practical battery reaching a steady state into the following multi-order training model equation: …, where represents the release voltage, represents the resting time, represents the steady-state voltage, and ~ and ~ represent multiple model training parameters; wherein the model training parameter set generated based on each specified remaining capacity percentage includes multiple model training parameters.

6. The battery remaining capacity percentage assessment method as described in claim 1 further includes the following steps: constructing a battery rebound voltage curve based on multiple estimated release voltages at multiple rebound time points, respectively, based on the candidate remaining capacity percentage selected each time.

7. The battery remaining power percentage assessment method as described in claim 1 further includes the following steps: constructing a tracking pattern based on the candidate remaining power percentage selected for the first time and its total estimated error value.

8. The battery remaining power percentage assessment method as described in claim 7 further includes the following steps: based on the candidate remaining power percentage selected in each subsequent selection other than the first one and its total estimated error value, moving the position of the tracking pattern or constructing a new tracking pattern.

9. The method for assessing the remaining battery charge percentage as described in claim 8 further includes the following step: marking the direction of movement of the tracking pattern.

Images