摘要
无机颜料的智能配色及光谱预测是颜料行业存在的重要问题,本文通过制备沉淀法制备Ca-FeOOH、ZnAl-LDHs-AG等两种黄色和蓝色颜料,通过XRD表征确认两种颜料的物相,并测试由Ca-FeOOH、ZnAl-LDHs-AG组成的物理混合得到不同比率的绿色颜料光谱。本文提出了一种基于混合K-M+DBN神经网络的混色机理建模方法,用于预测混色体系的反射率值,并对模型的可靠性进行验证。研究结果表明,模型预测准确率达到97%以上,具有优异的光谱预测能力。
The intelligent color matching and spectral prediction of inorganic pigments are significant issues in the pigment industry.In this paper,two yellow and blue pigments,namely Ca-FeOOH and ZnAl-LDHs-AG,were prepared by the precipitation method.The phases of the two pigments were confirmed by XRD characterization,and the spectra of green pigments obtained by physically mixing Ca-FeOOH and ZnAl-LDHs-AG at different ratios were tested.A color-mixing mechanism modeling method based on the hybrid K-M+DBN neural network was proposed to predict the reflectance values of the color-mixing system,and the reliability of the model was verified.The research results show that the prediction accuracy of the model is above 97%,demonstrating excellent spectral prediction capabilities.
作者
曹洺于
裘城聪
潘国祥
王远飞
肖春明
徐敏虹
李金花
CAO Minyu;QIU Chengcong;PAN Guoxiang;WANG yuanfei;XIAO Chunming;XU minhong;LI jinhua(Department of materials engineering,Huzhou university,Huzhou 313000;Zhejiang Huayuan Pigment Co.,Ltd,Deqing 313000)
出处
《光散射学报》
北大核心
2025年第4期856-863,共8页
The Journal of Light Scattering
基金
浙江省尖兵领雁研发攻关计划项目(2023C01112)
浙江省自然科学基金(LY21E040001)。
