Automated flame rectification process planning system in shipbuilding based on artificial intelligence

作者：Zhang, Zhiying; Jiang, Zhibin^*; Yu, Chengquan

来源：International Journal of Advanced Manufacturing Technology, 2006, 30(11-12): 1119-1125.

DOI：10.1007/s00170-005-0140-8

• automation
• flame rectification
• neural network
• rule-based reasoning

Distortion is a crucial problem with all welded fabrications in shipbuilding. Flame rectification is one of the most time-consuming processes. At present, it is still an experience-relying and manual operation. Because of lack in uniform flame rectification process, production has poor stability and productivity varies greatly among operators. In this paper, three typical kinds of welding components are chosen as study objects to generate the process planning of manual flame rectification automatically by artificial intelligence (AI) technology. In addition, we may also attain appropriate process parameters based on the model of neural network. A prototype system of automated flame rectification process planning is built. The system has been tested on actual components, and the results were found to be quite accurate and efficiency. Meanwhile, it establishes the fundament for the automation of flame rectification process.

• 出版日期2006-10
• 单位上海交通大学