Process monitoring and quality prediction system of GMAW robot welding based on digital twin

In the process of gas metal arc welding (GMAW), it is essential to ensure the quality of the weld seam. Traditional quality control methods suffer from latency and insufficient intelligence. To address these limitations, a process monitoring and quality prediction system of GMAW robot welding based on digital twin is proposed. Firstly, a five-dimensional model framework based on digital twin is proposed, and the construction process of the system is elaborated in detail. Then, a data management process based on the data-driven method is designed, and a digital twin model of robot welding is constructed, which integrates sensor data to achieve dynamic simulation and real-time process monitoring of robot welding. Besides, a quality prediction model based on BP neural network optimized by genetic algorithm is established, and the quality is quantified in combination with an evaluation mechanism. Finally, the GMAW system, grounded in the digital twin concept and integrating the physical platform with virtual model, has been developed and experimentally validated. The experimental results demonstrate that the system is highly practical, providing an effective solution for the real-time monitoring and prediction of welding quality.