JP4582583B2 - Secondary battery remaining capacity calculation method - Google Patents

Description

  The present invention relates to an improvement in a technique for detecting the internal state (particularly, the state of charge) of a vehicle power storage device using a neural network.

  For example, in a secondary battery such as a lead-acid battery, the correlation between the amount of electricity (for example, voltage, open circuit voltage, internal resistance, etc.), SOC, and remaining capacity (SOH) varies depending on the degree of deterioration. For this reason, there are problems that the detection accuracy of SOC and SOH deteriorates as the deterioration progresses, and variations in SOC and SOH for each battery, and the SOC and SOH of secondary batteries that are mass-produced are individually detected with high accuracy. It was considered difficult. For this reason, from the viewpoint of safety, there is also a problem that the usable charge / discharge range of the secondary battery must be set narrow including these variations.

  In order to improve this problem, a method of detecting SOC and SOH using a neural network that can flexibly cope with characteristic variations of the measurement target (hereinafter referred to as a neural network type battery state detection technique) has been proposed (Patent Literature). 1, 2).

  More specifically, Patent Document 1 proposes that the remaining capacity Te is detected by calculating with an already learned neural network using at least the open circuit voltage OCV, the voltage VO immediately after the start of discharge, and the internal resistance R as input parameters. .

Patent Document 2 introduces battery voltage, current, internal impedance (detection by an alternating current method) and temperature into a learned first neural network to calculate battery deterioration information, and this battery deterioration information and battery voltage, It has been proposed to calculate the remaining capacity of the battery by introducing current and internal impedance into a learned second neural network.
JP-A-9-243716 JP 2003-249271 A

However, the determination of SOC, SOH, etc. using the neural network type battery state detection technology according to Patent Documents 1 and 2 described above is much more burdensome on the circuit scale and computation scale than the remaining capacity detection method that does not perform neural network computation. However, there is a large error in the remaining capacity detection accuracy, and further improvement in detection accuracy is required for practical use. It has also been desired to improve the detection accuracy of the calculation method using the neural network while suppressing an increase in circuit scale and calculation scale.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a secondary battery remaining capacity calculation method capable of extracting excellent remaining capacity information by neural network calculation while avoiding an excessive calculation burden. It is said.

The method for calculating the remaining capacity of a secondary battery according to the present invention includes a calculation value obtained by a calculation using battery state data detected from a secondary battery or a storage state amount of the secondary battery as an input parameter. In the method for calculating the remaining capacity of the secondary battery calculated using a net, the input parameter includes a ratio to the voltage V of the secondary battery or a value when fully charged, an open circuit voltage Vo or a value when fully charged. Ratio of the internal resistance R or its value at full charge, a predetermined function value f (Vo, R) and a current I, and the function value f (Vo, R) is equal to the open circuit voltage Vo. The internal resistance R is an input variable and is a function value having a correlation with the current dischargeable current amount of the secondary battery.

  In this way, with respect to various secondary batteries with different degrees of battery deterioration and usage history, the remaining battery capacity calculation accuracy is good while suppressing the increase in the circuit scale and computation scale of the neural network arithmetic circuit. It was found that it can be improved.

In particular, in this method of calculating the remaining capacity of the secondary battery, the voltage / current history consisting of a large number of time-sequentially sampled pairs of voltage V and current I is not adopted as an input parameter of the neural network. And calculation processing load can be significantly reduced.

  More specifically, in the present invention, a function value f (Vo, R) having two parameters having different correlations between the battery deterioration and the remaining capacity, that is, the open circuit voltage Vo and the internal resistance R, respectively, is input to the neural network. Since the input variable is used, the influence of battery deterioration on the remaining capacity can be satisfactorily input as an input parameter to the neural network. As a result, it has been found that the remaining capacity can be calculated with high accuracy using appropriate input parameters while reducing the circuit scale and computation scale of the neural network.

  That is, a complex function value f (Vo, R) is calculated using a specific function value (open circuit voltage Vo, internal resistance R) obtained from the voltage history and current history as an input parameter, and this function value f (Vo, R) is calculated. ) Is used to calculate the remaining capacity of the neural network. The function value f (Vo, R) is preferably a function related to the current dischargeable current amount obtained from the open circuit voltage Vo and the internal resistance R. It was found that the function related to the current dischargeable current amount as an input parameter in the neural network calculation of the remaining capacity is very effective in improving the calculation accuracy.

  Note that the state of charge state mentioned above includes the remaining capacity (SOH) and the remaining capacity ratio (SOC). As the voltage (terminal voltage) and current of the secondary battery used as input parameters, it is preferable to extract the DC component with a low-pass filter for noise reduction or to calculate the average value in the immediately preceding predetermined period. It is. The open circuit voltage Vo and the internal resistance R can be calculated approximately from the past voltage / current data as usual.

  More specifically, since the dischargeable amount of the secondary battery varies depending on the degree of deterioration, and the degree of deterioration has a correlation with the open circuit voltage Vo and the internal resistance R, the influence of the degree of deterioration is taken into account in the calculation of the state of charge. It is considered preferable to use the voltage Vo and the internal resistance R as input parameters for the neural network operation. However, among the functions of the open circuit voltage Vo and the internal resistance R used in the present invention, the function value f (Vo, R) representing the dischargeable current of the secondary battery is further deteriorated than the open circuit voltage Vo and the internal resistance R. Since there is a stronger correlation with the state and the state of charge state (for example, SOC and SOH), by using this dischargeable amount correlation function value f (Vo, R) as an input parameter of the neural network operation, the processing load and circuit scale are increased. It is presumed that high-accuracy calculation was realized while suppressing the increase.

  Note that the voltage, the open circuit voltage Vo, and the internal resistance R as input parameters to the neural network include functions obtained by linearly converting them. For example, when K1 and K2 are constants, K1 · V + K2 may be used instead of the voltage V. The output error between the input parameter V and the input parameter (K1 · V + K2) can be easily converged by a neural network operation.

  The full charge ratio of the voltage V is a ratio with the voltage Vf at the time of full charge as a denominator and the current value of the voltage as a numerator, and the full charge ratio of the open circuit voltage Vo is an open circuit voltage at the time of full charge. Vof is the ratio with the current value of the open circuit voltage Vo as the numerator, and the full charge ratio of the internal resistance R is the ratio with the internal resistance R at the time of full charge as the denominator and the current value of the internal resistance R as the numerator. Say it. It has been found that by adopting such a full charge ratio, the detection accuracy is improved because comparison between different batteries becomes appropriate.

In a preferred embodiment, the function value f (Vo, R) is a function value having a correlation function value of an input variable ratio between at least the open circuit voltage Vo and internal resistance R (Vo / R). In this way, since this ratio (Vo / R) indicates the degree of battery deterioration, it is preferable that the ratio is suitable as an input parameter indicating a deterioration level correlated with the storage state amount in the neural network calculation of the storage state amount. all right.

  In a preferred aspect, the function value f (Vo, R) is a function value having a correlation with the function value Vo · Vo / R. Thereby, it turned out that the electrical storage state amount calculation precision can further be improved. The correlation function value f (Vo, R) can be converted into the following function by linearly converting Vo and R, respectively. Thereby, it turned out that the electrical storage state amount calculation precision can further be improved. k1 to k4 are constants. Preferably, k1 is 1, and k2 to k4 are 0.

f (Vo, R) = k1. ((Vo + k2). (Vo + k2) + k3) / (R + k4)
In a preferred embodiment, the function value f (Vo, R) is a function value (Vm−Vo) / R when Vm is a predetermined voltage. Thereby, it turned out that the electrical storage state amount calculation precision can further be improved. The function value f (Vo, R) can be converted into the following function by linearly converting Vo and R, respectively. k1 to k3 are constants. Preferably, k1 is 1, and k2 to k3 are 0.

f (Vo, R) = k1 · ((Vm−Vo) + k2) / (R + k3)
In a preferred embodiment, the predetermined voltage Vm is a discharge end voltage. Thereby, it turned out that the electrical storage state amount calculation precision can further be improved.

  In a preferred aspect, the predetermined function value f (Vo, R) is f (Vop, Rp) when the current value is the function value f (Vop, Rp) and the fully charged value is f (Vof, Rf). Vop, Rp) / f (Vof, Rf). It has been found that by adopting the full charge ratio in this way, comparison between different batteries becomes appropriate and detection accuracy is improved.

A calculation method using a neural network of a vehicle power storage device of the present invention will be specifically described with reference to the drawings.

(Circuit configuration)
A calculation method using the neural network of the vehicle power storage device of the first embodiment will be described below. First, the circuit configuration of the apparatus will be described with reference to the block diagram shown in FIG.

  101 is an in-vehicle power storage device (hereinafter also referred to as a battery), 102 is an in-vehicle generator that charges the in-vehicle power storage device, 103 is an electric device that forms an in-vehicle electric load fed from the in-vehicle power storage device 101, and 104 is an in-vehicle power storage device 101. A current sensor for detecting the charge / discharge current of the battery, 105 is a storage battery state detection device which is an electronic circuit device for detecting the state of the in-vehicle power storage device 101, 106 is a pre-processing circuit, and 107 is various inputs input from the pre-processing circuit 106. A neural network unit that outputs a predetermined storage state quantity (SOC in this embodiment) by performing a neural network operation on a parameter (input signal), and 108 is a power generation unit of the in-vehicle generator 102 based on a signal read from the neural network unit 107 or the like. A generator control device for controlling the amount. The pre-processing circuit 106 and the neural network unit 107 are realized by software calculation by a microcomputer device, but may be configured by a dedicated hardware circuit.

  The pre-processing circuit 106 simultaneously samples and reads a pair of the voltage V of the in-vehicle power storage device 101 and the current I from the current sensor 104 at regular time intervals and stores them for a predetermined period, and based on these voltage / current pairs, will be described later. The neural network unit 107 calculates predetermined function values (average voltage Va, average current Ia, open circuit voltage Vo, internal resistance R, dischargeable current amount correlation function value Vo · Vo / R) used as input parameters. Va is an average value of the voltage V of the secondary battery 101 in the latest predetermined period, and Ia is an average value of the charge / discharge current I of the secondary battery 101 in the latest predetermined period. Such an average value calculation can be easily performed. A low-pass filter output may be used for the calculation of Va and Ia.

  A method of calculating the open circuit voltage Vo and the internal resistance R will be described with reference to FIG. FIG. 2 is a voltage-current distribution diagram showing a two-dimensional distribution of pairs of average voltage Va and average current Ia sampled and stored at a predetermined interval. From each voltage-current pair, a linear approximation formula L indicating the relationship between the average voltage Va and the current Ia is calculated and generated by the least square method, and the intercept (open circuit voltage Vo) and / or the slope (internal resistance R) is calculated by this linear approximation formula. Is calculated each time a pair of the average voltage Va and the average current Ia is input, and the open circuit voltage Vo and the internal resistance R are calculated. Since the creation of a linear approximation formula using this kind of least squares method and the extraction of the open circuit voltage using this linear approximation formula are known matters, further explanation is omitted. Thereafter, a dischargeable current amount correlation function value f (Vo, R) = Vo.Vo / R = Pm is calculated as a new input parameter adopted this time. This Pm is a function having a positive correlation with the current dischargeable power.

  Next, the full charge ratio of the average voltage Va, the open circuit voltage Vo, the internal resistance R, and the function value f (Vo, R) is calculated. These full charge ratios refer to the ratio of the current value to the average voltage Va at full charge, the open circuit voltage Vo at full charge, the internal resistance R, and the function value f (Vo, R). Specifically, the full charge average voltage Vaf which is the value of the average voltage Va at the time of the latest full charge, the full charge open circuit voltage Vof which is the value of the open circuit voltage Vo at the time of the latest full charge, the internal at the time of the latest full charge The full charge internal resistance Rf, which is the value of the resistance R, and the full charge function value f (Vof, Rf), which is the value of the function value f (Vo, Rf) at the time of the latest full charge, are calculated at the time of the latest full charge, It is remembered. Using these values and the values calculated this time, the ratio Va / Vaf, ratio Vo / Vof, ratio R / Rf, ratio (VoVoR) / (VofVof / Rf) are input to the neural network calculation. Calculated as a parameter.

  The full charge determination for calculating the full charge average voltage Vaf, the full charge open circuit voltage Vof, the full charge internal resistance Rf, and the full charge function value f (Vof, Rf) will be described below with reference to FIG. . FIG. 3 is the same as the voltage-current distribution diagram used in FIG. 2, and shows a two-dimensional distribution of pairs of average voltage Va and average current Ia sampled and stored at a predetermined interval. When the coordinate point specified by the pair of the detected average voltage Va and average current Ia enters the full charge region shown in FIG. 3, it is determined that the battery is fully charged, and the value of the average voltage Va at this time is determined as the full charge average voltage Vaf. Assuming that the value of the open circuit voltage Vo at this time is the full charge open circuit voltage Vof and the internal resistance R at this time is the full charge internal resistance Rf, the function value f (Vo, R) = Vof · Vof / Rf at this time is It is stored as a full charge function value f (Vof, Rf). These memories are rewritten every time the full charge determination is made.

  The ratio Va / Vaf, ratio Vo / Vof, ratio R / Rf, ratio (Vo · Vo / R) / (Vof · Vof / Rf) and current I thus obtained are input as input parameters to the neural network. The remaining capacity rate (SOC) is calculated by operating the neural network, and the calculated remaining capacity rate (SOC) is output. FIG. 4 shows a flow of the neural network calculation of the remaining capacity ratio (SOC) described so far. The neural network unit 107 performs a neural network operation on the remaining capacity ratio (SOC) using the five input parameters (r1 to r4 and the average current Ia), and outputs the calculated remaining capacity ratio (SOC) to the outside. A suitable input parameter other than these five input parameters may be used as appropriate.

  Next, the above-described neural network operation will be described with reference to the block diagram shown in FIG. The learned neural network unit 107 is a three-layer feedforward type that learns by the error back propagation method, but is not limited to this type. The neural network unit 107 includes an input layer 201, an intermediate layer 202, and an output layer 203. However, the neural network unit 107 is actually configured by software processing executed at a predetermined calculation interval. That is, since the neural network unit 107 is actually configured by software operation of a microcomputer circuit, the circuit configuration shown in FIG. 1 is only functional.

  The input layer 201 includes five input cells in this embodiment, but the number of input cells may be further increased. Each input cell individually receives the input parameters (r1 to r4, Ia), and outputs them as input data (input signals) to all the operation cells of the intermediate layer 202. Each computation cell in the intermediate layer 202 performs a neural network computation described later on each input data input from each input cell in the input layer 201, and outputs the computation result to an output cell in the output layer 203. The output cell of the output layer 203 outputs a charge rate (SOC) in this embodiment.

  The learning of the neural network unit 107 shown in FIG. 5 will be described below.

If the input data of the jth cell of the input layer 201 of the neural network unit 107 is INj, and the coupling coefficient of the jth cell of the input layer 201 and the kth cell of the intermediate layer 202 is Wjk, the input data to the kth cell of the intermediate layer The input signal is
INPUTk (t) ＝ Σ (Wjk * INj) (j = 1 to 2m + 3)
It becomes. The output signal from the kth cell of the intermediate layer is
OUTk (t) = f (x) = f (INPUTk (t) + b)
It is represented by b is a constant. f (INPUTk (t) + b) is a nonlinear function called a sigmoid function with INPUTk (t) + b as an input variable.
f (INPUTk (t) + b) = 1 / (1 + exp (-(INPUTk (t) + b)))
Is a function defined by If the coupling coefficient between the kth cell of the intermediate layer 202 and the cell of the output layer 203 is Wk, the input signal to the output layer is
INPUTo (t) = Σ Wk * OUTk (t)
k = 1 to Q
It is represented by Q is the number of cells in the intermediate layer 202. The output signal at time t is
OUT (t) = L * INPUTo (t)
It becomes. L is a linear constant.

  In this specification, the learning process means that each cell has a minimum error between a final output OUT (t) at time t and a teacher signal (that is, a true value tar (t)) measured in advance. Is to optimize the coupling coefficient between. The output OUT (t) is an output parameter to be output by the output layer 203, and here is the SOC at time t.

  Next, a method for updating each coupling coefficient will be described.

The update of the coupling coefficient Wk between the kth cell in the intermediate layer and the cell in the output layer is
Wk = Wk + △ Wk
Done in Here, ΔWk is defined as follows.

△ Wk = −η * ∂Ek / ∂Wk η; Constant
= η * [OUT (t) − tar (t)] * [∂OUT (t) / ∂Wk]
= η * [OUT (t) − tar (t)] * L * [∂INPUTo (t) / ∂Wk]
= η * L * [OUT (t) − tar (t)] * OUTk (t)
It is represented by Ek is an amount representing an error between teacher data and network output, and is defined by the following equation.

Ek = [OUT (t)-tar (t)] x [OUT (t)-tar (t)] / 2
Next, an update rule for the coupling coefficient Wjk between the kth cell in the intermediate layer 202 and the jth cell in the input layer 201 will be described. The update of the coupling coefficient Wjk is realized by the following equation.

Wjk = Wjk + △ Wjk
Here, ΔWjk is defined as follows.

△ Wjk = -η * ∂Ek / ∂Wjk
= -Η * [∂Ek / ∂INPUTk (t)] * [∂INPUTk (t) / ∂Wjk]
= -Η * [∂Ek / ∂OUTk (t)] * [∂OUTk (t) / ∂INPUTk (t)] * INj
= -Η * [∂Ek / ∂OUT (t)] * [∂OUT (t) / ∂INPUTo] *
[∂INPUTo / OUTk (t)] * f '(INPUTk (t) + b) * INj
= −η * (OUT (t) −tar (t)) * L * Wk * f ′ (INPUTk (t) + b) * INj
= −η * L * Wk * INj * (OUTsoc (t) −tar (t)) * f ′ (INPUTk (t) + b)
Here, f ′ (INPUTk (t) + b) is a differential value of the transfer function f.

  The output OUT (t), that is, the SOC at time t is calculated again with the new coupling coefficients Wk and Wjk updated in this manner, and the coupling coefficient is continuously updated until the error function Ek becomes a predetermined minute value or less. In this way, the neural network unit 107 performs learning by updating the coupling coefficient so that the error function Ek becomes a predetermined value or less.

  A flowchart of the learning process is shown in FIG. However, the state of charge as an output parameter of the power storage device to be output by the neural network unit 107 is SOC (charge rate), but may be a remaining capacity (SOH).

  First, an appropriate initial value of each coupling coefficient of the neural network unit 107 is set (step 302). This may be determined appropriately using, for example, a random number. Next, an input signal for learning is individually input to each cell of the input layer 201 of the neural network unit 107 (step 303), and this input signal is subjected to a neural network calculation using the initial value of the coupling coefficient described above. The SOC as an output parameter is calculated (step 304). Next, the error function Ek is calculated by the above method (step 305), and it is determined whether or not this error function is smaller than a predetermined minute value th (step 306). If the error function Ek is larger than the minute value th, the update amount ΔW of each coupling coefficient defined in the learning process is calculated (step 307), and each coupling coefficient is updated (step 308). Next, the learning input signal is input again to each cell of the input layer 201 to calculate the SOC. Next, the miscalculation function Ek is evaluated, and if it is below the minute value th, it is determined that the learning has been completed (step 309), and this learning process is terminated. If the error function Ek is not below the minute value, the coupling coefficient is updated again, SOC calculation is performed, the error function Ek is evaluated, and this process is repeated until the error function Ek falls below the minute value.

  If the neural network unit 107 is made to learn by executing the learning process described above before shipment of the product for several types of batteries each having a representative charge / discharge pattern, Regardless of the manufacturing variation of the in-vehicle battery, the SOC can be calculated with high accuracy by the neural network calculation of the SOC of the in-vehicle storage battery that is traveling thereafter.

(Test results)
Actually, the current and terminal voltage during 10.15 mode driving were measured with five batteries (see Fig. 7) with different capacities and deterioration levels, the above five types of input signals of the neural network were calculated, and the true value of the SOC (current Learning was performed using a teacher signal as calculated from the integrated value.

  Next, the SOC of a new deteriorated battery was calculated using this learned neural network, and compared with the current integration result. The results are shown in FIGS. 8 to 10, the SOCs of the three test batteries were subjected to neural network calculation using a total of five input parameters described above with the ratio (Vo · Vo / R) / (Vof · Vof / Rf) as input parameters. FIG. 11 to FIG. 13 show the SOC of the same three test batteries, and the total of four other previously described without using the ratio (Vo · Vo / R) / (Vof · Vof / Rf) as an input parameter. The result of the neural network operation using the input parameters is shown. It has been found from FIGS. 8 to 13 that the SOC calculation accuracy can be improved by simply adding this ratio as an input parameter.

  Furthermore, the correlation between the ratio Vo / Vof and the SOC was examined for nine types of batteries. The result is shown in FIG. The correlation was 0.86. Next, for nine types of batteries, the correlation between the power ratio (Vo · Vo / R) / (Vof · Vof / Rf) and the SOC was examined. The result is shown in FIG. The correlation was 0.93. From FIG. 15, it is estimated that the SOC calculation accuracy is improved by using the above power ratio having a more excellent correlation as an input parameter instead of simply using the open circuit voltage Vo as an input parameter for the neural calculation as in the prior art. Is done.

A calculation method using the neural network of the vehicle power storage device of the second embodiment will be described below. The circuit configuration and the calculation method are essentially the same as those in the first embodiment. However, as the input parameter, the function value f (Vo, R) to be added is the current discharge instead of the full charge ratio of (Vo · Vo / R). The difference is that a full charge ratio of (Vm−Vo) / R equal to the possible current Im is adopted. Vm is a predetermined discharge end voltage value, and is 10.5 V in this embodiment. FIG. 2 shows the relationship between the discharge end voltage value Vm and the dischargeable current Im.

(Test results)
Actually, the current and terminal voltage during 10.15 mode driving were measured with five batteries (see Fig. 7) with different capacities and deterioration levels, the above five types of input signals of the neural network were calculated, and the true value of the SOC (current Learning was performed using a teacher signal as calculated from the integrated value.

  Next, the SOC of a new deteriorated battery was calculated using this learned neural network, and compared with the current integration result. The results are shown in FIGS. 16 to 18 use the SOC of the three test batteries, using a total of five input parameters described above with the ratio ((Vm−Vo) / R) / (((Vm−Vof) / Rf) as input parameters. FIG. 19 to FIG. 21 show the SOCs of the same three test batteries using the ratio ((Vm−Vo) / R) / ((Vm−Vof) / Rf) as input parameters. 6 shows the result of the neural network calculation using the other four input parameters already described, and it is possible to improve the SOC calculation accuracy by simply adding the ratio as an input parameter from FIGS. found.

(Modification)
The ratio ((Vm−Vo) / R) / ((Vm−Vof) / Rf) and the ratio (Vo · Vo / R) / (Vof · Vof / Rf) may be simultaneously set as input parameters. In this case, since the correlation level between the input parameter and the SOC is further improved, one or both of the open circuit voltage Vo and the internal resistance R may be omitted from the input parameter.

1 is a block diagram illustrating a circuit configuration of a device according to Embodiment 1. FIG. It is a figure which shows the example of the approximate expression for calculating an open circuit voltage and internal resistance in Example 1. FIG. It is a figure which shows the full charge area | region for the full charge determination in Example 1. FIG. 3 is a flowchart illustrating an SOC calculation process in the first embodiment. It is a block diagram which shows the structure of the neural network part which comprises a battery state detection apparatus. It is a flowchart of the learning process of the neural network part of FIG. It is a characteristic view of the battery used for learning. It is a figure which shows the SOC calculation result of the 1st test battery of Example 1. It is a figure which shows the SOC calculation result of the 2nd test battery of Example 1. It is a figure which shows the SOC calculation result of the 3rd test battery of Example 1. It is a figure which shows the reference SOC calculation result of the 1st test battery of Example 1 (when not using the full charge ratio of function value Vo * Vo / R). It is a figure which shows the reference SOC calculation result of the 2nd test battery of Example 1 (when not using the full charge ratio of function value Vo * Vo / R). It is a figure which shows the reference SOC calculation result of the 3rd test battery of Example 1 (when not using the full charge ratio of function value Vo * Vo / R). It is a characteristic view which shows the correlation with the full charge ratio of the open circuit voltage Vo, and SOC. It is a characteristic view which shows the correlation with the full charge ratio of function value Vo * Vo / R and SOC. It is a figure which shows the SOC calculation result of the 1st test battery of Example 2. It is a figure which shows the SOC calculation result of the 2nd test battery of Example 2. It is a figure which shows the SOC calculation result of the 3rd test battery of Example 2. It is a figure which shows the reference SOC calculation result of the 1st test battery of Example 2 (when not using the full charge ratio of (Vm-Vo) / R). It is a figure which shows the reference SOC calculation result of the 2nd test battery of Example 2 (when not using the full charge ratio of (Vm-Vo) / R). It is a figure which shows the reference SOC calculation result of the 3rd test battery of Example 2 (when not using the full charge ratio of (Vm-Vo) / R).

Explanation of symbols

101 on-vehicle power storage device 102 on-vehicle generator 104 current sensor 105 storage battery state detection device (calculation means)
106 Pre-processing circuit unit 107 Neural network unit (neural network unit)
108 generator control device 109 correction signal generation unit 201 input layer 202 intermediate layer 203 output layer

Claims (6)

Remaining capacity calculation of the secondary battery that uses a neural network to calculate the storage state quantity of the secondary battery using the calculated value obtained by calculation using the battery status data detected from the secondary battery or the battery status data as an input parameter In the method
The input parameters are the ratio of the voltage V of the secondary battery or its full charge value, the ratio of the open circuit voltage Vo or its full charge value, the ratio of the internal resistance R or its full charge value. , At least a predetermined function value f (Vo, R) and current I,
The function value f (Vo, R) is a function value having a correlation with the current dischargeable current amount of the secondary battery using the open circuit voltage Vo and the internal resistance R as input variables. Battery remaining capacity calculation method . In the secondary battery remaining capacity calculation method according to claim 1,
The function value f (Vo, R), the secondary battery which is a function value having a correlation function value of an input variable ratio between at least the open circuit voltage Vo and internal resistance R (Vo / R) Remaining capacity calculation method . In the secondary battery remaining capacity calculation method according to claim 2,
The function value f (Vo, R) is the residual capacity calculation method of the secondary battery, which is a function value having a correlation function value Vo · Vo / R. In the secondary battery remaining capacity calculation method according to claim 2,
The function value f (Vo, R), when the Vm and the predetermined voltage, the remaining capacity calculation method of a secondary battery characterized in that it is a function value (Vm-Vo) / R. In the secondary battery remaining capacity calculation method according to claim 4 ,
The method for calculating a remaining capacity of a secondary battery, wherein the predetermined voltage Vm is an end-of-discharge voltage. In the secondary battery remaining capacity calculation method according to any one of claims 1 to 5,
The predetermined function value f (Vo, R) is f (Vop, Rp) / when the current value is the function value f (Vop, Rp) and the fully charged value is f (Vof, Rf). A method for calculating a remaining capacity of a secondary battery, wherein f (Vof, Rf).

Images