3D Printer State Prediction: A Deep Learning Model Approach

In the process of 3D printing, the devices used to print usually encounter errors and problems that are not easily detected by the device operator. Undetected errors can cause serious damage to the 3D printer and lead to the output being counted as reject, thus leading to both lost time and resources. The research focuses on the development of a system to monitor the process of 3D printing and predict the future values of the temperature of the printer. This research aims to evaluate the various deep learning models such as the multilayer perceptron (MLP), long short term memory (LSTM), and integration of both LSTM and convolutional neural networks (CNN). Furthermore, a comparative analysis on what works best in predicting temperature values will be performed. The system shall allow the collection of data from various sensors attached to a 3D printer through a web-based application. In testing the model, forecasting metrics such as the root mean square error, mean average error, mean absolute percentage error, and the r-square values will be evaluated. After a thorough comparative analysis, the application of LSTM performs best.