研究目的
To improve product quality while maintaining a cost-effective manufacturing environment through the application of Artificial Intelligence.
研究成果
The implementation of A.I. in LDM manufacturing has led to quality control improvement, reduction of human work, efficient usage of big data, and the prospect of future automation. The development time was less compared to building conventional image recognition algorithms under commercial software.
研究不足
The study is limited to the application of A.I. in laser diode module manufacturing and may not be directly applicable to other manufacturing processes without adaptation.
1：Experimental Design and Method Selection:
Implementation of A.I. using machine learning and deep learning for data analysis and classification problems in laser diode module manufacturing.
2：Sample Selection and Data Sources:
Data from laser diode module manufacturing processes.
3：List of Experimental Equipment and Materials:
Laser diode modules, optical fibers, Scikit-learn, TensorFlow, NVIDIA GeForce GTX1060, Raspberry Pi.
4：Experimental Procedures and Operational Workflow:
Application of Scikit-learn for LD selection and LDM allocation, Convolutional Neural Network for fiber stripped-edge inspection, and Autoencoder for fiber facet inspection.
5：Data Analysis Methods:
Machine learning and deep learning techniques including Lasso linear regression, CNN, and Autoencoder.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
