激光冲击区表面质量的人工神经网络研究

大量的实验表明,经激光冲击处理后,材料受冲击区的表面质量与材料的疲劳寿命有着明显的关系。因此,表面质量是判断激光冲击强化效果的重要手段。将人工神经网络技术用于激光冲击处理后试件的表面质量分析,建立了激光参数与激光冲击处理后试件的表面质量之间的联系,并用其实现了对冲击处理后的试件表面质量的预测。研究及实验表明,该方法不仅具有准确及稳定性好等特点,而且这种预测能力在实际应用中还具有不断提高的智能特性。

Study of the surface qualities of laser shock-processing zones using an artificial neural network

A lot of experiments have shown that there is an obvious relation between surface qualities of specimen after laser shock-processing(LSP) and its fatigue life.Consequently,the LSP effects can be evaluated by surface qualities in LSP areas.In this paper,an artificial neural network(ANN) is utilized to study the surface qualities of specimen after LSP.Based on the data obtained in the experiment,an ANN is established.The trained ANN could acquire the relations between surface qualities and laser parmeters.From the verification of aluminium alloy 2024-T62,it is proved that the neural network can successfully predict the surface quality grades of specimen after LSP,and easily determine the laser parameters under different production conditions.The research and experimental results show that the ANN has not only the accuracy and good stability,but also the intelligent improving control ability during process.