Classification of Autoclave Temperature via Deep Learning

In the composite material processing, autoclave forming is a commonly-used approach by the action of heat and pressure at the same time. The temperature distribution could greatly affect the quality of composite material. However, high temperature and cacuum leakage could result in poor quality of composite products. It is important to discover the reasons that caused undesirable composite product during the processing. In recent years, deep learning technique has achieved great success in improving manufacturing processing. In this paper, we applied CNN and long short term memory (LSTM) models for analysis the processing temperature types of the composite materials. In this study, we have made a comparative analysis of two different classification algorithms with 8 categories autoclave. The results show that CNN model was able to correctly recognize eight types of the autoclave in 83.33%, and 72.22% accuracy of LSTM model. With this intelligence models, which make it possible to perform in the autoclave forming processing to trace out the types of composite processing temperature.