Prediction of OLED temperature distribution based on a neural network model

The market share of organic light-emitting device (OLED) screens has increased significantly in recent years. With traditional liquid crystal display (LCD) screen, the positive effects of OLED screens with bright colors and lower power consumption are indicated. LCD screens demand a backlight LED as a light source, while other OLED pixels can emit light within their own. As a result, the OLED's power consumption is reduced, and its thickness is reduced. Therefore, OLED screens are becoming dramatically popular. Although there are numerous benefits of OLED screens, it also contains some of certain drawbacks. After a usage period, the OLED brightness deteriorates, and the luminance loss is diluted with the different regions of the OLED panel, named OLED burn-in. Although there have been some studies on OLED luminance degradation, including research on OLED luminance deterioration under various temperature circumstances, none of them have addressed the impacts of temperature in different sections of a panel. The temperature forecast for different sections of an OLED panel is discussed in this study. Besides, the four temperature sensors positioned at the back of the OLED panel, as well as the images displayed on the screen at the time are used to anticipate the temperature of the panel. The prediction method is a neural network, which uses previously acquired data to train the model. Following completion of the training, the temperature distribution of the OLED panel is predicted using the four temperature sensors and the input images. The mean square error used to calculate the prediction error is less than 0.65 °C.