Details

Autor(en) / Beteiligte

• van Driel, Willem Dirk
• Yazdan Mehr, Maryam

Titel

Artificial Intelligence and LED Degradation

Ist Teil von

• Reliability of Organic Compounds in Microelectronics and Optoelectronics, 2022, p.297-316

Ort / Verlag

Switzerland: Springer International Publishing AG

Erscheinungsjahr

2022

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Artificial Intelligence (AI) is intelligence performed by designed algorithms. Recently, there are many advancements in the AI field that benefited by the improvements in computational power from the hardware. Meanwhile, there is easy access and availability of data from various platforms. AI technologies with data science can help solve many industrial problems like tracking Solid-state Lighting (SSL) system degradation, classifying LED failures, and predicting Remaining Useful Life (RUL) for the system. The modern AI technologies and fields include Deep Learning (DL) and Computer Vision (CV). Deep Convolutional Neural Network CNN) and Recurrent Neural Network (RNN) is the fundamental network that can be easily adapted to the imaging system for material degradation analysis in the LEDs. In this chapter, firstly, it introduces representative supervised learning algorithms for LED degradation monitoring and failure classification. Then, it gives a brief introduction to Artificial Neural Network (ANN) and the latest development of the networks in Deep Learning. Also, it talks about how to choose, design, and test the model for the imaging problems in the LED applications.