Study of on-line temperature monitoring system for hull welding based on Xgboost-PSO
Jiaqi Qu, Zheng Qian · 2019
Welding is a very complicated process in ship manufacturing engineering. Lack of temperature control will lead to various defects in welding. Therefore, temperature monitoring during welding is of great significance to the quality of ship welding. This research is based on the temperature data collected by thermocouples, and then communicates with the host computer through Wi-Fi. The machine learning algorithm Xgboost is used to realize the non-linear correction of thermocouples and the cold-end compensation in the host computer, and the particle swarm optimization (PSO) algorithm is used to construct dynamic compensator to reduce dynamic error. Finally, the on-line monitoring of welding temperature is realized in the form of host computer software.