基于人工神经网络的焊缝宽度预测

Weld width prediction based on artificial neural network

摘要: 研究了用神经网络预测焊缝宽度的方法。首先对焊接质量检测系统的一些相关问题进行了研究,考虑了弧焊特性的提取、焊接质量的预报以及人工神经网络模型(ANN)的应用,设计了一个基于人工神经网络的焊接质量检测系统,给出系统的组成结构,ANN被用于预测焊缝宽度,建立了焊缝宽度预测的人工神经网络模型。为了验证建立的ANN模型的可行性,进行了仿真研究。仿真结果表明,所建立的ANN模型可预测焊缝宽度,基于人工神经网络的焊接质量检测系统是有效的。

Abstract: The method of weld width prediction based on artificial neural network(ANN) was studied,and the weld quality could be estimated by it.Problems and questions relating to the welding process monitoring system were considered,such as picking up the characteristics of arc welding,welding quality prediction and the application of ANN.The welding quality monitor system based on ANN was designed and its composing structure was given.The weld quality prediction model based artificial neural network is build up,and simulation is given.The results of simulation showed that the ANN model could predict the weld width,and the welding quality monitor system based on ANN was efficient.