摘要
初步探讨了微秒激光辐射不锈钢表面的着色机理,在激光焦距及填充间距一定的条件下,研究了激光扫描速度、加工次数及激光重复频率对304不锈钢着色效果的影响,并在此基础上通过建立三个并联的Elman神经网络,研究了激光扫描速度及激光加工次数与着色块色度(H)、饱和度(S)、亮度(B)之间的非线性关系。使用该神经网络在给定的激光参数下进行HSB值预测,预测曲线与真实值曲线吻合良好,其中色度的测试平均相对误差为4.04%,饱和度的测试平均相对误差为13.33%,亮度的测试平均相对误差为4.05%。所建立的神经网络模型具有良好的预测精度,实际加工图案颜色与预测颜色具有较高的一致性。
Objective Laser metal coloring technology has the advantages of pollution-free,simple operation,and high degree of automation,and has broad application prospects in the field of metal coloring.Various colors can be produced on stainless steel,copper,aluminum,titanium,and other metal surfaces.The laser coloring mechanism on metal surfaces includes the color of the oxide layer itself,the film interference effect produced by the oxide layer,and the structural color induced by laser irradiation.The existing theory cannot perfectly explain the surface coloring mechanism.The quantitative relationship between the laser processing parameters and the coloring effect cannot be established,which results in poor repeatability of laser coloring processing and hinders the application of laser coloring technology.With the development of artificial intelligence,artificial neural networks can fit and predict the data relationship of complex nonlinear systems.Elman neural network is one of the main neural networks,which is a typical dynamic recurrent neural network.Elman neural network has good global stability and strong computing ability.In this paper,the laser coloring mechanism is discussed in detail.Experimental data are used to establish an Elman neural network prediction model to determine the numerical relationship between the laser processing parameters and the hue,saturation and brightness(HSB)values.Methods An ultraviolet microsecond laser(355nm)is used to irradiate the surface of 304 stainless steel to obtain laser induced colors.Under the fixed pulse duration,focal length,and fill spacing,the 304 stainless steel surface is processed by changing repetition rate(20--130kHz),scanning speed(150--700 mm/s),and the number of laser processing cycles(1--12).The surface morphologies and elemental compositions of the color patches are analyzed by using scanning electron microscope(SEM)and X-ray energy spectrometer.Coloring mechanism is discussed based on three kinds of coloring mechanisms:structural coloring,film interference e
作者
张龙达
李好发
安丰硕
王志文
郑宏宇
Zhang Longda;Li Haofa;An Fengshuo;Wang Zhiwen;Zheng Hongyu(School of Mechanical Engineering,Shandong University of Technology,Zibo,Shandong 255000,China)
出处
《中国激光》
EI
CAS
CSCD
北大核心
2022年第8期103-111,共9页
Chinese Journal of Lasers
基金
淄博市校城融合项目(2019ZBXC087,2019ZBXC168)。
