Search

With the development of electronic technology, the requirements for electronic components are gradually becoming chip, miniaturization and high precision. In this case, problems such as scattered sample parts and poor detection accuracy and stability in traditional production testing methods seriously affect production quality. Therefore, in-depth study of intelligent material discharge detection has become an important way to solve the problem. In this paper, the needs of intelligent detection are comprehensively analyzed, and the machine vision imaging optimization is deeply studied, feature matching and small sample capture placement technologies are learned. In order to realize intelligent discharge detection of micro scattered equipment, an automatic discharge system is developed, which integrates counting module, defect detection module and automatic discharge module.

This paper deeply analyzed the types of information that can be mined in component reliability data. According to specific needs, the clustering measurement method of reliability data sample similarity is studied, the distance measurement strategy is obtained, the algorithm steps and analysis methods of Apriori association rule data mining are improved, the reliability data prediction and analysis methods based on statistical methods are compared. AR prediction model, MA model and other component reliability information time prediction methods have formed data mining software such as clustering, correlation analysis and prediction, discovered a large amount of hidden reliability information, and realized the comprehensive mining and utilization of electronic component reliability data.

In order to solve the problems such as unreasonable plan, opaque process, insufficient flexibility and excessive manual participation in the process of screening, identification and quality consistency inspection of electronic components in China, a software and hardware construction scheme of intelligent detection system for electronic components is proposed, which integrates the cutting-edge technologies such as machine vision, deep learning, IOT(internet of things), data mining. Relying on the National Semiconductor Device Quality Supervision and Inspection Center, part of the construction work has been completed, and the information, automation and intelligence of the testing business have been realized.

Aiming at the problems of low efficiency and low accuracy caused by manual defect detection of electronic components, FCN, SegNet/DeconvNet and DeepLab and other deep learning technologies are studied. Using Caffe, Keras, PyTorch and other frameworks, an automatic defect identification system for electronic components is developed to distinguish qualified products from unqualified ones, so as to realize intelligent defect detection of electronic components. At the same time, the detection accuracy rate is greatly improved, and the quality of electronic components is guaranteed.

Aiming at the importance of failure analysis of electronic components, the complexity of the knowledge system and the high dependence of experts, this paper conducts in-depth research on the application of expert systems. According to the characteristics of component failure analysis, the reasoning fault tree analysis engine is innovatively designed, and the active knowledge assistant technology is introduced to assist analysts to obtain failure mechanism information in real time. The low threshold and high output of failure analysis of electronic components are realized, which fills the gap at home and abroad.

Due to the increasingly thin, chip, miniaturized and high-precision electronic components, the samples in many links of the production and testing process of components are scattered and disordered. Meanwhile, the low efficiency of manual testing and the instability of detection accuracy make the intelligent material discharging detection of components extremely urgent. In this paper, intelligent requirements for detection are comprehensively analyzed, and technologies such as machine vision imaging optimization, deep learning feature matching, and tiny sample grasping and placement are deeply studied. An automatic discharging system integrating counting module, defect detection module and automatic discharging module is developed to realize intelligent discharging detection of tiny bulk disordered devices.