| During the actual welding production, many welding procedure qualifications have to be performed. In order to replace some the welding procedure qualifications, Artificial Neural Network technology and Genetic Algorithms together were applied to prediction of the mechanical property of the welded joints.Based on a great deal of welding procedure qualification data, the mechanical properties prediction models of welding joints were established according to base-metal composition, filler metal composition, welding parameters, post heat treatment and tested temperatures by using neural network These models can predict yield strength, ultimate tensile strength, elongation, reduction ratio of area of alloy steels welded with Gas Metal Arc welding, Submerged Arc welding and Gas Tungsten Arc welding.In addition, Genetic Algorithm was used to optimize the Back-propagation neural network connection weights and improve the models' predicted precision and generalization ability. The performance analysis shows that the predicted trend agrees well with the previous research work and the predicted error is less than 5%. It is obvious that the models will be more applicable and valuable in the practice with the enlargement of database and the data-covering space.A welding procedure management system for different standards based on Client/Server structure was developed by applying database technology and expert system technology. The system can be used by multi-users with different levels of authorization at the same time,The knowledge base was built through taking 5 general criterions into consideration, including American Society of Mechanical Engineers IX, American Welding Society D1.1, American Petroleum Institute, American Bridge Welding Criterion and British-European Criterion. Finally, data were dealt with safely due to the security mechanism of DBMS (database management system) and its own user management module. |
PDF Full Text Request