Big data-based battery inspection method
The present invention relates to a big data-based battery inspection method. Specifically, the present invention is for providing a big data-based battery inspection method which applies information on a metal part of a battery measured by an eddy current sensor to a discrimination function so as to quickly inspect the state of the metal part.
1. A method for inspecting a battery based on big data comprising:
an eddy current induction step of inputting an input current that is an AC current into a transmission coil and irradiating a primary magnetic field generated in the transmission coil to a metal portion of the battery to induce an eddy current in the metal portion of the battery;
an output voltage measurement step of inputting a secondary magnetic field generated by the eddy current generated in the eddy current induction step into a receiving coil and measuring an induced electromotive force in the receiving coil generated by the secondary magnetic field;
an impedance calculation step of calculating an impedance based on the input current value and the output voltage value;
an impedance analysis step of separating a real part and an imaginary part from the impedance value, inputting them into a discriminant function and outputting a discrimination value from the discriminant function; and
a condition judgement step of judging a condition of the metal portion of the battery based on the discrimination value output in the impedance analysis step,
wherein
in the eddy current induction step, the input current comprises a plurality of partial AC currents having different set frequencies,
in the impedance calculation step, the impedance is calculated by dividing the output voltage value by the input current value, and the impedance comprises a plurality of partial impedances corresponding to each of the set frequencies,
in the impedance analysis step, a plurality of the discriminant functions are provided, and each of the discrimination values for the plurality of discriminant functions is calculated by Equation 4 below:
D
m
=
∑
n
=
1
N
(
Cr
m
,
n
R
n
+
Cx
m
,
n
X
n
)
[
Equation
4
]
D m is a discrimination value of the m th discriminant function, N is the number of the set frequencies, R n is a real part of the partial impedance corresponding to the n th set frequency, X n is an imaginary part of the partial impedance corresponding to the n th set frequency, and Cr m,n and Cx m,n are discrimination coefficients corresponding to the n th set frequency of the m th discriminant function.
2. The method for inspecting a battery based on big data according to claim 1 , wherein the input current is calculated by Equation 1 below:
I
=
∑
n
=
1
N
I
Pn
e
2
π
f
n
ti
[
Equation
1
]
I is an input current, N is the number of partial AC current values, I Pn is a complex number representing the amplitude and phase of the n th partial AC current, f n is a set frequency of the n th partial AC current, and t is a time.
3. The method for inspecting a battery based on big data according to claim 1 , wherein in the eddy current induction step,
the transmission coil irradiates the primary magnetic field to the surface of the metal portion of the battery while a longitudinal direction of the transmission coil is perpendicular to the surface of the metal portion of the battery.
4. The method for inspecting a battery based on big data according to claim 1 , wherein in the output voltage measurement step,
the receiving coil is disposed such that a longitudinal direction of the receiving coil is perpendicular to the surface of the metal portion of the battery.
5. The method for inspecting a battery based on big data according to claim 1 , wherein the discrimination coefficients are calculated based on sample impedance values measured in metal portions of a plurality of the sample batteries and condition information values of the metal portions of the plurality of the sample batteries.
6. The method for inspecting a battery based on big data according to claim 5 , wherein the condition information value comprises at least one information of an ohmic resistance value, a welding thickness, and a tensile strength of the metal portion of the sample battery.
7. The method for inspecting a battery based on big data according to claim 1 , wherein in the impedance analysis step, m discrimination values are calculated for each of the metal portion of the battery, and
in the condition judgement step, the condition of the metal portion of the battery is judged by concurrently considering at least two discrimination values among the m discrimination values.
8. The method for inspecting a battery based on big data according to claim 7 , wherein the condition judgement step comprises:
a discrimination value selection step of selecting two discrimination values among the m discrimination values,
a graph output step of outputting a two-dimensional discrimination graph in which the discrimination values selected in the discrimination value selection step are used as each axis, and
a data marking step of displaying the discrimination value of the metal portion of the battery as a coordinate on the discrimination graph output in the graph output step.
9. The method for inspecting a battery based on big data according to claim 8 , wherein in the graph output step,
a plurality of the condition areas indicating the condition of the metal portion of the battery are displayed on the discrimination graph.