JP7849598B2 - Battery life estimation device and battery life estimation method - Google Patents

Description

This disclosure relates to a battery life estimation device and a battery life estimation method.

The introduction of stationary battery storage systems is expanding, aimed at the efficient use of renewable energy and power supply during disasters. Since battery performance gradually degrades, understanding the battery's lifespan is crucial for the smooth operation of the system. Therefore, battery life estimation is being carried out.

One example of a battery life estimation technique is the use of the so-called square root rule, which utilizes the fact that the square root of the battery's operating time is proportional to its capacity (see, for example, Patent Document 1).

Japanese Patent Publication No. 2013-254710

Incidentally, the battery life estimation technology described in Patent Document 1 aims to more accurately grasp the performance degradation of the battery, particularly towards the end of its lifespan, based on the performance degradation status of the positive and negative electrodes of the battery. As this example illustrates, improving the accuracy of battery life estimation is one of the requirements. The inventors of this invention were seeking a life estimation method different from the above-mentioned improvement measures.

[1] A battery life estimation device that solves the above problem is a battery life estimation device that estimates the life of a battery based on an approximate formula obtained from measured values of the battery state of the battery, and comprises: a period calculation unit that calculates the cycle time for charge-discharge cycles in which charge-discharge operations were performed during the operating period of the battery, and the storage time for storage periods in which no charge-discharge operations were performed; and a life estimation unit that obtains an approximate formula of a power law using least-squares approximation based on the measured values of the battery state of the battery, the cycle time, and the storage time, and calculates life information up to the desired life of the battery based on the approximate formula.

According to the above configuration, the cycle time related to charge-discharge cycles during the battery's operating period and the storage time related to storage are calculated separately. Next, based on the measured battery state values, cycle time, and storage time, an approximate formula using a power law based on least-squares approximation is obtained. Then, based on the obtained approximate formula, the battery's lifespan information up to the desired lifespan is calculated. In other words, by focusing on the significant difference in battery performance between charge-discharge cycles and simple storage, and performing calculations separately, a significant improvement in lifespan estimation accuracy can be expected. Furthermore, by using a power law for the approximation formula, it is possible to perform approximations other than the square root rule, which is commonly used in battery lifespan estimation, depending on the situation. This also contributes to the expected improvement in lifespan estimation accuracy.

[2] The battery life estimation device described in [1] above includes an estimation feasibility determination unit that determines whether the number of measured battery state values of the battery is equal to or greater than a determination value for determining whether the number is sufficient for estimating the battery life, and allows the battery life estimation process, including the life estimation unit, based on the determination that it is equal to or greater than the determination value.

According to the above configuration, battery life estimation is performed when a sufficient number of measured battery state values are above the specified threshold. This further improves the accuracy of the life estimation.

[3] In the battery life estimation device described in [1] or [2] above, the life estimation unit obtains an estimated value of the battery state value of the battery based on the approximation formula, and includes an estimation accuracy determination unit that determines whether the estimated value of the battery state value of the battery is in an approximation pattern where the correlation coefficient with the measured value of the battery state value of the battery is equal to or greater than a determination value, and allows the battery life estimation process, including the life estimation unit, to proceed based on the determination that it is equal to or greater than the determination value.

According to the above configuration, battery life estimation is performed when the correlation coefficient between the estimated battery state value and the measured value is greater than or equal to a certain threshold. This further improves the accuracy of life estimation.

[4] In the battery life estimation device described in any one of [1] to [3] above, the life estimation unit obtains an estimated value of the battery state of the battery based on the approximation formula, and is configured to obtain the approximation formula with the condition that the latest measured value of the battery state of the battery matches the corresponding estimated value.

According to the above configuration, an approximation formula is obtained, including the condition that the latest measured battery state value of the storage battery matches the corresponding estimated value, and the storage battery's lifespan is estimated. Matching the latest measured value with the estimated value leads to an improvement in the accuracy of the lifespan estimation.

[5] In the battery life estimation device described in any one of [1] to [4] above, the life estimation unit is configured to acquire the approximate formula excluding the measured values of the battery state values of the battery during the initial stages of operation.

According to the above configuration, the measured values of the battery's state during the initial stages of operation are excluded, and an approximate formula is obtained to estimate the battery's lifespan. Excluding older measured values from the initial stages of operation leads to an improvement in the accuracy of the lifespan estimation.

[6] A battery life estimation method for solving the above problems is a battery life estimation method that estimates the battery life based on an approximate formula obtained from measured values of the battery state value of the battery, and calculates the cycle time for charge-discharge cycles in which charge-discharge operations were performed during the battery's operating period, and the storage time for storage periods in which no charge-discharge operations were performed, and obtains an approximate formula of a power law using least-squares approximation based on the measured values of the battery state value, the cycle time, and the storage time, and calculates life information up to the desired life of the battery based on the approximate formula.

With the above configuration, a significant improvement in the accuracy of battery life estimation can be expected, similar to the battery life estimation device described above.

The battery life estimation device and battery life estimation method described herein offer a significant improvement in life estimation accuracy.

This is a diagram showing the configuration of a storage battery and a lifespan estimation device in one embodiment. This is a flowchart illustrating the battery life estimation process in the same embodiment. This is an explanatory diagram relating to the estimation of the battery life in the same embodiment. This is an explanatory diagram relating to the estimation of the battery life in the same embodiment. This is an explanatory diagram regarding the estimation of battery life in the modified example. This is an explanatory diagram regarding the estimation of battery life in a comparative example.

Below, an embodiment of a battery life estimation device and a battery life estimation method will be described with reference to the drawings.
(Overall configuration of the lifespan estimation device 20 that estimates the lifespan of the storage battery 10)
Figure 1 shows the overall configuration of this embodiment, including the battery 10 and the life estimation device 20. The battery 10 is, for example, a stationary lithium-ion battery. The battery 10 is installed for purposes such as the effective utilization of renewable energy such as solar power generation and power supply during disasters. The battery 10 is electrically connected to the life estimation device 20. In this embodiment, the battery condition of the battery 10 is diagnosed by the life estimation device 20, and its lifespan is estimated.

The life estimation device 20 comprises a battery state diagnosis unit 30, an estimation feasibility determination unit 40, a period calculation unit 50, a life estimation unit 60, an estimation accuracy calculation unit 70, an estimation accuracy determination unit 80, and a storage unit 90 related to the storage battery 10. The storage unit 90 stores battery state data 91, period data 92, relational expression data 93, life estimation data 94, and determination data 95, respectively, in an input/output manner. Note that the arrows shown in Figure 1 represent the main data flow, and some are omitted.

(Battery status diagnosis unit 30)
As shown in Figures 1 and 4, the battery state diagnostic unit 30 measures the battery capacity Q as the current battery state value of the battery 10 in operation. The measurement of the battery capacity Q is performed periodically at predetermined intervals. In this embodiment, the periodic measurement is set, for example, once a week at a predetermined time. The battery capacity Q includes the actual measured value itself, or the ratio of the actual measured value to the total capacity (referred to as the capacity retention rate in this embodiment). In Figure 4, etc., the battery capacity Q is shown as the capacity retention rate. In this embodiment, the battery capacity Q is used to measure the battery state value of the battery 10, but other parameters correlated with the battery state value of the battery 10, such as the internal resistance of the battery 10, can also be used. The actual measured value _Qx of the battery capacity Q, which is measured periodically by the battery state diagnostic unit 30, is sequentially stored and accumulated in the storage unit 90 as battery state data 91, associated with time information.

(Estimation possibility determination unit 40)
As shown in Figures 1, 3, and 4, the estimation feasibility determination unit 40 determines whether to perform a lifespan estimation of the battery 10. That is, the estimation feasibility determination unit 40 determines whether there is sufficient measured value _Qx data to be used for estimating the lifespan of the battery 10, by determining whether the number of battery capacity Q data stored in the storage unit 90 is equal to or greater than the determination value N _a . The number of battery capacity Q data is the number during a predetermined period from the start time _T0 of the battery 10 lifespan estimation to the current time T _now in this embodiment. In this embodiment, the determination value N _a is set to, for example, 5 [points]. It is believed that at least 5 [points] of battery capacity Q data are necessary to ensure sufficient accuracy in estimating the lifespan of the battery 10. The determination value N _a is stored in the storage unit 90 as determination data 95.

(Period calculation unit 50)
The period calculation unit 50 calculates the total time t _sum , cycle time t _cyc , storage time t _st , and time ratio A related to the battery 10 during the operating period. The total time t _sum is the total time from time T ₀ to time T _now . The cycle time t _cyc is the time related to the charge-discharge cycle in which the battery 10 is charged and discharged. The storage time t _st is the time simply spent in storage when the battery 10 is not charged or discharged. The total time t _sum is the sum of the cycle time t _cyc and the storage time t _st , and can be expressed as t _sum = t _cyc + t _st . The time ratio A is the ratio of the cycle time t _cyc to the storage time t _st , and can be expressed as A = t _st / t _cyc .

The period calculation unit 50 measures the cycle time t _cyc related to the charge-discharge cycle when the input/output current I of the battery 10 in operation changes by a determination value ±I _a (see Figure 3). In this embodiment, the determination value ±I _a is set to, for example, ±1 [A]. On the other hand, when the input/output current I of the battery 10 changes by less than the determination value ±I _a, it measures the storage time t _st related to the mere storage period. The cycle time t _cyc and storage time t _st that occur at each time are accumulated. Incidentally, the total time t _sum , cycle time t _cyc , and storage time t _st may be measured individually, but any two of them may be measured and the remaining one calculated. The total time t _sum , cycle time t _cyc , storage time t _st , and time ratio A are stored in the storage unit 90 as period data 92.

(Life estimation section 60)
The life estimation unit 60 obtains an approximate power law formula [Equation 1] using least squares approximation based on the data of battery capacity Q, cycle time t _cyc , and storage time t _st within the time range from the above- _mentioned time T ₀ to the current time T now. In other words, the coefficients Q ₀ , k ₁ , k ₂ , α ₁ , and α ₂ of the approximate formula are calculated. In addition, along with the calculation of the coefficients Q ₀ , k ₁ , k ₂ , α ₁ , and α ₂ , the estimated value Q _a of the battery capacity Q is also calculated.

The above equation [Equation 1] is equivalent to [Estimated battery capacity Q _a ] = [Coefficient _Q0 (initial value of battery capacity Q)] - [Power of cycle time t _cyc with coefficients _k1 and _α1 ] - [Power of storage time t _st with coefficients _k2 and _α2 ]. The above equation [Equation 1] is stored in the memory unit 90 as relational equation data 93. Various equations related to life estimation, such as equations [Equation 2] to equation [Equation 5] described later, are also stored in the memory unit 90 as relational equation data 93.

The above formula [Equation 1] used to estimate the lifespan of the battery 10 in this embodiment is structured by dividing it into a term for the cycle time t _cyc related to the charge-discharge cycle of the battery 10 and a term for the storage time t _st related to the mere storage period. This is because the inventors believe that the performance degradation of the battery 10 related to the charge-discharge cycle is different from the performance degradation related to the mere storage period. Furthermore, the formula is structured as a so-called root law, which utilizes the fact that the square root of the operating time of the battery 10 (including the time itself or the number of cycles) is proportional to the battery capacity, that is, _a "power law" in which the coefficients _{α1 and α2} are not only "1/2" but also other real numbers. This is because the inventors believe that this can provide a better approximation than the root law depending on the condition of the battery 10. The estimated value _Qa of the battery capacity Q calculated in this way is also sequentially stored in the storage unit 90 as lifespan estimation data 94.

(Estimated accuracy calculation unit 70)
The estimation accuracy calculation unit 70 calculates the correlation coefficient r between the estimated value _Qa and the measured value _Qx of the battery capacity Q.

(Estimation accuracy determination unit 80)
The estimation accuracy determination unit 80 determines whether the correlation coefficient r is greater than or equal to the determination value r _a , that is, whether the estimated value Q _a approximates the measured value Q _x to the desired extent. In this embodiment, the determination value r _a is set to, for example, "0.95". In this embodiment, the determination value r _a is stored in the storage unit 90 as determination data 95. If the correlation coefficient r is less than the determination value r _a , the estimation accuracy determination unit 80 determines that the approximation of the estimated value Q _a of the battery capacity Q calculated by the life estimation unit 60 to the measured value Q _x is less than desired, and determines that the estimation accuracy is poor. On the other hand, if the correlation coefficient r is greater than or equal to the determination value r _a , the estimation accuracy determination unit 80 determines that the approximation of the estimated value Q _a of the battery capacity Q calculated by the life estimation unit 60 to the measured value Q _x is better than desired, and determines that the estimation accuracy is good.

If the estimation accuracy is determined to be poor, the life estimation unit 60 changes the time range from time _T0 to time _T0 ^' where the next data is available, and recalculates the correlation coefficient r between the estimated value _Qa and the measured value _Qx of the battery capacity Q for the time range from the changed time _T0 ^' to time _Tnow . In other words, the oldest data is removed. Then, the above judgment and calculation operations are repeated with the aim of making the correlation coefficient r equal to or greater than the judgment value _ra , that is, determining that the estimation accuracy is good. As this is repeated, older data is sequentially removed. Of course, the number of data points for the battery capacity Q is kept within a range that is equal to or greater than the judgment value _Na .

(Various calculation processes of the life estimation unit 60)
Here, substituting the time ratio A = t _st / t _cyc , that is, t _st = A・t _cyc , into the above formula [Equation 1] used in the lifetime estimation unit 60, we obtain the following formula [Equation 2].

In calculating the estimated cycle life t _{cyc_end} of the storage battery 10, the above equation [Equation 2] becomes the following equation [Equation 3].

If the life determination capacity Q _end of the storage battery 10 is set to, for example, 60% in terms of capacity retention rate, the coefficients _Q0 , _k1 , _k2 , _α1 , and _α2 related to life estimation can be obtained from the above formula [Equation 1], and the estimated cycle life t _{cyc_end} of the storage battery 10 can be obtained from the above formula [Equation 3]. That is, the life estimation unit 60 calculates the estimated cycle life t _{cyc_end} using the above formula [Equation 3].

Since the total time t _sum is given by t _sum = t _cyc + t _st , the estimated lifetime t _{sum_end} from time T ₀ (or possibly time T ₀ ^' ) to time T _end when the lifetime determination capacity Q _end is reached is given by the following equation [Equation 4].

Furthermore, by rearranging the above equation [Equation 4] using the time ratio A = t _st / t _cyc , that is, t _st = A・t _cyc , and substituting the estimated cycle lifetime t _{cyc_end} obtained from the above equation [Equation 3], it is possible to obtain the estimated lifetime t _{sum_end} . In other words, the lifetime estimation unit 60 calculates the estimated lifetime t _{sum_end} using the above equation [Equation 4].

The estimated lifespan t _{sum_end} obtained from equation [4] above can be used to obtain the total remaining time t _{sum_rem} of the battery 10 by subtracting the operating time t _{sum_now} from time _T0 to the current time T _now , as shown in equation [5].

The life estimation unit 60 calculates the total remaining time t _{sum_rem} of the battery 10 using the above formula [Equation 5]. The life estimation device 20 then displays the calculation results, such as the total remaining time t _{sum_rem} of the battery 10, on a display device (not shown) or outputs the results to the relevant equipment, and terminates the life estimation process.

(Processing procedure for estimating the lifespan of the storage battery 10)
The battery life estimation process for the storage battery 10 in this embodiment is performed according to the flow shown in Figure 2.
In step S1, the battery capacity Q is measured as the current battery state value of the storage battery 10. The battery capacity Q is measured periodically. The battery capacity Q measured each time is used to estimate the lifespan of the storage battery 10 and is stored in the storage unit 90 in association with the time information of the measurement.

In step S2, it is determined whether the number of data entries for the battery capacity Q stored in the memory unit 90 is equal to or greater than the determination value N _a . It is determined whether there is sufficient data to ensure the accuracy of the battery life estimation for the battery 10. This process proceeds to step S3 only if the number of data entries for the battery capacity Q is equal to or greater than the determination value N _a .

In step S3, the cycle time t _cyc related to the charge-discharge cycle of the battery 10, the storage time t _st related to the simple storage process, the total time t _sum related to the operation of the battery 10, and the time ratio A are calculated.

In step S4, the coefficients _Q0 , _k1 , _k2 , _α1 , _α2 of the above formula [Equation 1] related to the current lifespan estimation of the storage battery 10, and the estimated value _Qa of the storage battery capacity Q are calculated using the least squares approximation with the above formula [Equation 1].

In step S5, the correlation coefficient r between the estimated value _Qa and the measured value _Qx of the battery capacity is calculated.
In step S6, it is determined whether the correlation coefficient r is greater than or equal to the judgment value r _a , that is, whether the estimated value Q _a approximates the measured value Q _x to the desired extent. If the correlation coefficient r is less than the judgment value r _a , that is, if the approximation of the estimated battery capacity Q _a to the measured value Q _x is less than desired, the estimation accuracy is determined to be poor. This process proceeds to step S7.

In step S7, the time range is changed from time _T0 to time _T0 ^' where the next data is available, and steps S2 to S6 are repeated. In other words, the correlation coefficient r between the estimated value _Qa and the measured value _Qx of the battery capacity is recalculated while excluding the oldest data, and the process is repeated with the aim of determining that the correlation coefficient r is equal to or greater than the judgment value _ra , i.e., that the estimation accuracy is good. Note that the inclusion of step S2 in this repeated process ensures that there are enough data points for battery capacity Q that are equal to or greater than the judgment value _Na .

In step S6 above, if the correlation coefficient r is greater than or equal to the judgment value r _a , that is, if the approximation of the estimated battery capacity Q _a with the measured value Q _x is better than desired, the estimation accuracy is judged to be good. This process proceeds to step S8.

In step S8, the estimated cycle life t _{cyc_end} of the battery 10 is calculated using the above formula [Equation 3]. The time required for charge-discharge cycles until the battery 10 reaches its life determination capacity Q _end is calculated.

In step S9, the estimated lifespan t _{sum_end} of the battery 10 is calculated using the above formula [Equation 4]. The total time until the battery 10 reaches its lifespan determination capacity Q _end is calculated.
In step S10, the total remaining time _{tsum_rem} of the battery 10 is calculated by subtracting the operating time _{tsum_now} from the estimated lifespan _{tsum_end} using the above formula [Equation 5]. In other words, the total remaining time _{tsum_rem} until the battery 10 reaches its lifespan determination capacity Q _end can be determined, and this can be used to consider maintenance, replacement, etc. of the battery 10.

(Operation of this embodiment)
The operation of this embodiment will now be described.
In this embodiment, the lifespan estimation of the battery 10 is performed using a calculation method that includes the least-squares approximation expressed by the above formula [Equation 1], which is expected to significantly improve the accuracy of the remaining total time t _{sum_rem} . The formula [Equation 1] used in this embodiment separates the term for the cycle time t _cyc related to the charge-discharge cycle of the battery 10 from the term for the storage time t _st related to the storage period. Since the performance of the battery 10 differs greatly between the charge-discharge cycle and the simple storage period, we believe that performing the calculations separately will significantly contribute to improving the accuracy of the lifespan estimation. Furthermore, formula [Equation 1] is designed to allow not only the square root rule, which is commonly used in the lifespan estimation of the battery 10, but also approximations other than the square root rule. We believe that this also contributes to improving the accuracy of the lifespan estimation.

As shown in Figure 4, using the life estimation method of this embodiment, the estimated value Qa and the measured value _Qx of the battery capacity Q of the battery 10 are sufficiently similar throughout the entire period from the beginning of operation to the end of life. In particular, although the estimated value _Qa and the measured value _Qx tend to diverge at the end of life when the battery ₁₀ reaches the life determination capacity Q _end , in this embodiment the error time _ta between them is kept sufficiently small.

In contrast, in the comparative example's life estimation method, which uses a general square root rule that does not differentiate the operating time of the battery 10, as shown in Figure 6, although the estimated value _Qa and the measured value _Qx of the battery capacity Q of the battery 10 approximate each other in the middle of operation, this embodiment is still superior. Furthermore, a slight discrepancy is observed between the estimated value _Qa and the measured value _Qx in the early stages of operation. In particular, at the end of life when the life determination capacity _Qend is reached, the discrepancy between the estimated value _Qa and the measured value _Qx is very large, and the error time _tx between them is very large.

Thus, compared to the comparative example, the battery life estimation method of this embodiment can be used to significantly improve the accuracy of battery life estimation, especially in the end-of-life stage where accuracy tends to deteriorate.

(Effects of this embodiment)
The effects of this embodiment will now be explained.
(1) The cycle time t _cyc related to the charge-discharge cycle during the operation period of the storage battery 10 and the storage time t _st related to the storage period are calculated separately. Next, based on the measured value _Qx of the storage battery capacity Q used as the battery state value of the storage battery 10, the cycle time t _cyc , and the storage time t _st , an approximate formula of the power law using least squares approximation (the approximate formula shown in [Equation 1] above) is obtained. Then, based on the obtained approximate formula, life information up to the life determination capacity Q _end , which is the desired lifespan of the storage battery 10 (in this embodiment, the estimated cycle life t _{cyc_end} , estimated life t _{sum_end} , and remaining total time t _{sum_rem} of the storage battery 10) is calculated. In other words, by focusing on the fact that the performance of the storage battery 10 differs greatly between the charge-discharge cycle and the simple storage period, and proceeding with the calculations separately, it is possible to expect a significant improvement in the accuracy of life estimation. Furthermore, by using a power law for the approximate formula, it is possible to perform approximations other than the square root law, depending on the situation, in addition to the square root law that is commonly used in the life estimation of the storage battery 10. This also allows us to expect an improvement in the accuracy of lifespan estimation.

(2) If there are a sufficient number of measured values _Qx of the battery capacity Q used as the battery state value of the storage battery 10 that are equal to or greater than the judgment value N _a , the process proceeds from step S2 in Figure 2 to step S3 and beyond, and the lifespan of the storage battery 10 is estimated. This also allows for further improvement in the accuracy of lifespan estimation.

(3) When the correlation coefficient r between the estimated value Qa of the battery capacity _Q used as the battery state value of the battery 10 and the measured value _Qx is in an approximate state of being greater than _or equal to the judgment value ra, the process proceeds from step S6 to step S7 and beyond in Figure 2, and the lifespan of the battery 10 is estimated. In this embodiment, the lifespan is estimated by sequentially removing older data starting from the initial time _T0 of operation of the battery 10. This also allows for further improvement in the accuracy of the lifespan estimation.

(Example of change)
This embodiment can be implemented with the following modifications. This embodiment and the following modifications can be combined with each other to the extent that they do not contradict each other technically.

The numerical values and formulas mentioned above are examples only and may be changed as appropriate.
The operating time of the battery 10 includes either the time itself or the number of cycles.
- The total operating time t _sum , cycle time t _cyc , and storage time t _st of the battery 10 may be calculated by individual timing, or any two of them may be timing and the remaining one calculated.

• Although the battery capacity Q was used as the battery state value of the battery 10 for calculating the lifespan, other parameters correlated with the battery state value of the battery 10, such as the internal resistance of the battery 10, may also be used.

- In calculating the lifetime estimate, although not specifically mentioned in the above embodiment, as shown in Figure 5, it may be a condition that the measured value _Qx at time T _now , i.e., the latest measured value _Qx , matches the estimated value _Qa . By matching the latest measured value _Qx with the estimated value _Qa in this way, the accuracy of the lifetime estimate can be improved.

- While the lifespan estimation was previously based on data acquired from time _T0 to the current time _Tnow , as shown in Figure 5, the lifespan estimation of the battery 10 may also be performed by excluding older data from time _T0 at the beginning of the battery 10's operation to, for example, time _Ta . Time _Ta may be set at a fixed interval from the start of the battery 10's operation, or it may be a time that fluctuates in conjunction with the current time _Tnow . By excluding older data from the early stages of operation, which have a relatively different pattern of change compared to the end of the battery 10's lifespan, it is possible to further reduce the error time _tb between the estimated value _Qa and the measured value _Qx , especially at the end of the lifespan. In this way, the accuracy of the lifespan estimation can be improved.

- As lifespan information up to the desired lifespan, the estimated cycle life t _{cyc_end} , estimated lifespan t _{sum_end} , and remaining total time t _{sum_rem} of the storage battery 10 may all be calculated and displayed, or at least one of them may be calculated and displayed. In addition, lifespan information may be expressed in a manner other than time.

The storage battery 10 may have a configuration other than a stationary lithium-ion storage battery.
The battery 10 may also be used in conjunction with renewable energy power generation equipment other than solar power generation.

(Note)
The technical concepts that can be understood from the above embodiments and modified examples are described below.
(i) The approximate formula for the power law is:
The estimated battery state value is calculated as follows: [Initial battery state value] - [Power of battery cycle time] - [Power of battery storage time].
Battery life estimation device.

(b) The approximate formula for the power law is:
The estimated battery state value is calculated as follows: [Initial battery state value] - [Power of battery cycle time] - [Power of battery storage time].
Method for estimating the lifespan of a storage battery.

10...Battery 20...Life Estimation Device 30...Battery Status Diagnosis Unit 40...Estimation Feasibility Determination Unit 50...Period Calculation Unit 60...Life Estimation Unit 70...Estimation Accuracy Calculation Unit 80...Estimation Accuracy Determination Unit 90...Storage Unit 91...Battery Status Data 92...Period Data 93...Relational Expression Data 94...Life Estimation Data 95...Determination Data Q...Battery Capacity (Battery Status Value)
_Qx ... Measured value _Qa ... Estimated value t _cyc ... Cycle time t _st ... Storage time t _{cyc_end} ... Estimated cycle lifetime (lifetime information)
t _{sum_end} …Estimated lifespan (lifespan information)
t _{sum_rem} ... Total remaining time (lifespan information)
r...correlation coefficient r _a ...decision value N _a ...decision value

Claims (6)

A battery life estimation device that estimates the life of a storage battery based on an approximate formula obtained from measured values of the battery state of the storage battery,
A period calculation unit calculates the cycle time for charge-discharge cycles in which charge-discharge operations were performed during the operating period of the storage battery, and the storage time for storage periods in which no charge-discharge operations were performed.
The system comprises: a life estimation unit that obtains an approximate formula for a power law based on the measured battery state value of the storage battery, the cycle time, and the storage time using least-squares approximation, and calculates life information up to the desired life of the storage battery based on the approximate formula;
Battery life estimation device. The system includes an estimation feasibility determination unit that determines whether the number of measured battery state values of the storage battery is equal to or greater than a determination value for determining whether the number is sufficient for estimating the lifespan of the storage battery, and that, based on the determination that it is equal to or greater than the determination value, allows the storage battery life estimation process, including the lifespan estimation unit, to proceed.
The battery life estimation device according to claim 1. The life estimation unit obtains an estimated value of the battery state value of the storage battery based on the approximation formula,
The system includes an estimation accuracy determination unit that determines whether the estimated value of the battery state value of the storage battery is in an approximate state where the correlation coefficient with the measured value of the battery state value of the storage battery is greater than or equal to a determination value, and that permits the battery life estimation process, including the life estimation unit, based on the determination that it is greater than or equal to the determination value.
The battery life estimation device according to claim 1. The life estimation unit obtains an estimated value of the battery state value of the storage battery based on the approximation formula,
The system is configured to obtain the approximation formula, with the condition that the latest measured value of the battery state of the storage battery matches the corresponding estimated value.
The battery life estimation device according to claim 1. The life estimation unit is configured to obtain the approximate formula by excluding the measured values of the battery state values of the storage battery during the initial stages of operation.
The battery life estimation device according to claim 1. A method for estimating the lifespan of a storage battery, which estimates the lifespan of the storage battery based on an approximate formula obtained from measured values of the battery state of the storage battery,
The cycle time for charge-discharge cycles during the operation period of the aforementioned battery, and the storage time for storage periods during which no charge-discharge operations were performed are calculated.
Based on the measured battery state value of the storage battery, the cycle time, and the storage time, an approximate formula for a power law is obtained using least-squares approximation, and life information up to the desired lifespan of the storage battery is calculated based on the approximate formula.
Method for estimating the lifespan of a storage battery.