Method for diagnosing quality of aluminum-based copper-clad plate

• Main Authors: SONG XIAOHUI, WANG JIANYE, ZHUANG CHUNSHENG, ZHANG HONGMIN, ZHAO LANPU, WU SHUNLI
• Format: Patent
• Language: Chinese; English
• Published: 09.03.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention provides a method for diagnosing the quality of an aluminum-based copper-clad plate, and the method comprises the steps: constructing a quick quality diagnosis model correlated with themain quality characteristics of the aluminum-based copper-clad plate through continuous learning and model optimization based on the high-definition image characteristics of the surface of the aluminum-based copper-clad plate before and after the solidification of heat-conducting insulating glue through a deep learning network; and rapidly predicting the quality of the aluminum-based copper-clad plate before lamination, and providing a basis for further process optimization. The method can be used for on-line rapid quality diagnosis, and plays an important promoting role in improving the yieldand reducing the manufacturing cost. 本发明提供了一种铝基覆铜板质量的诊断方法，其基于铝基覆铜板的导热绝缘胶固化前后表面的高清图像特征，利用深度学习网络，通过持续学习与模型优化，构建了与铝基覆铜板主要质量特征关联的快速质量诊断模型，实现在压合前对铝基覆铜板质量进行快速预测，为进一步工艺优化提供依据。该方法能够在线快速质量诊断，对于提高成品率，降低制造成本具有重要促进作用。