Method of predicting lifetime of display device
A method of predicting a lifetime of a display device according to an embodiment includes creating a machine learning model based on prior degradation rate data according to a degradation time for each of pixels, measuring a first degradation rate data for each of the pixels by inputting a voltage to each of the pixels, predicting a second degradation rate data for each of the pixels using the machine learning model, and estimating a degradation rate for each of the pixels according to a degradation time based on the first degradation rate data and the second degradation rate data.
1 . A method of predicting a lifetime of a display device, the method comprising:
creating a machine learning model based on prior degradation rate data according to a degradation time for each of pixels;
measuring, during a first period, a first degradation rate data for each of the pixels by inputting a voltage to each of the pixels;
predicting, during a second period, a second degradation rate data for each of the pixels using the machine learning model, the machine learning model predicting degradation amount information of a driving element for each of the pixels and degradation amount information of a light emitting device (LED) of each pixel to produce the second degradation rate data; and
estimating a degradation rate for each of the pixels according to a degradation time based on the first degradation rate data and the second degradation rate data,
wherein an end time of the first period and a start time of the second period are the same,
wherein each of the pixels includes:
a light emitting device which emits light; and
a driving element providing a driving current to the light emitting device, and in the estimating of the degradation rate for each of the pixels, the degradation rate is estimated by modeling degradation amount information of the light emitting device and a degradation amount information of the driving element with a degradation model defined as a degradation rate function over time,
wherein the degradation model is expressed by Equation 2 below,
L
(
t
)
L
(
0
)
=
[
1
+
k
×
{
1
-
e
-
1
×
(
t
ε
)
γ
}
]
×
e
-
1
×
(
t
τ
)
β
,
[
Equation
2
]
and
in the Equation 2, L(t) is a current luminance, L(0) is an initial luminance, each of τ and ε is a parameter which determines a rate of luminance decrease, each of β and γ is a parameter which determines a form of luminance decrease, and t is time for which luminance decrease proceeded.
2 . The method of claim 1 , wherein
the measuring of the first degradation rate data is performed in a first non-zero period having a first time length, and
the predicting of the second degradation rate data is performed in a second period having a second time length.
3 . The method of claim 2 , wherein the first time length of the first period and the second time length of the second period are same.
4 . The method of claim 2 , wherein the first time length of the first period is shorter than the second time length of the second period.
5 . The method of claim 1 , wherein the machine learning model is created based on Linear Regression, Polynomial Regression, Principal Components Regression, Partial Least Squares Regression, Random Forest, Gradient Boosting, Extreme Gradient Boosting, Ridge Regression, and/or Lasso Regression.
6 . The method of claim 1 , wherein the machine learning model is created based on Multilayer Perceptron, Bayesian Neural Networks, Radial Basis Functions, Generalized Regression Neural Networks, K-Nearest Neighbor Regression, Classification And Regression Tree, Support Vector Regression, and/or Gaussian Processes.
7 . The method of claim 1 , wherein the light emitting device includes an organic material.
8 . The method of claim 1 , wherein the measuring the first degradation rate data comprises measuring each pixel multiple times.
9 . The method of claim 2 , wherein the measuring the first degradation rate data comprises measuring each pixel multiple times throughout the first period.
10 . The method of claim 2 , wherein a sum of the first time length and the second time length correspond to the lifetime of the display device.
11 . A method of predicting a lifetime of a display device, the method comprising:
creating a machine learning model based on prior degradation rate data according to a degradation time for each of pixels;
measuring a first degradation rate data for each of the pixels by inputting a voltage to each of the pixels;
predicting a second degradation rate data for each of the pixels using the machine learning model; and
estimating a degradation rate for each of the pixels according to a degradation time based on the first degradation rate data and the second degradation rate data, wherein
each of the pixels includes:
a light emitting device which emits light; and
a driving element providing a driving current to the light emitting device, and
in the estimating of the degradation rate for each of the pixels, the degradation rate is estimated by modeling degradation amount information of the light emitting device and a degradation amount information of the driving element with a degradation model defined as a degradation rate function over time, wherein
the degradation model is expressed by Equation 2 below,
L
(
t
)
L
(
0
)
=
[
1
+
k
×
{
1
-
e
-
1
×
(
t
ε
)
γ
}
]
×
e
-
1
×
(
t
τ
)
β
,
[
Equation
2
]
and
in the Equation 2, L(t) is a current luminance, L(0) is an initial luminance, each of τ and ε is a parameter which determines a rate of luminance decrease, each of β and γ is a parameter which determines a form of luminance decrease, and t is time for which luminance decrease proceeded.
12 . The method of claim 11 , wherein the light emitting device includes an organic material.
13 . The method of claim 11 , wherein the measuring the first degradation rate data comprises measuring each pixel multiple times.
14 . The method of claim 11 , wherein
the measuring of the first degradation rate data is performed in a first non-zero period having a first time length, and
the predicting of the second degradation rate data is performed in a second period having a second time length.
15 . The method of claim 14 , wherein the measuring the first degradation rate data comprises measuring each pixel multiple times throughout the first time period.
16 . The method of claim 11 , wherein the machine learning model is created based on Linear Regression, Polynomial Regression, Principal Components Regression, Partial Least Squares Regression, Random Forest, Gradient Boosting, Extreme Gradient Boosting, Ridge Regression, and/or Lasso Regression.
17 . The method of claim 11 , wherein the machine learning model is created based on Multilayer Perceptron, Bayesian Neural Networks, Radial Basis Functions, Generalized Regression Neural Networks, K-Nearest Neighbor Regression, Classification And Regression Tree, Support Vector Regression, and/or Gaussian Processes.