基于吸光度与物质浓度的相关关系建立了一种PET薄膜印刷的混合油墨组分预测模型。 该方法首先依据光的传播理论和光在墨层与承印材料中的多重内反射原理, 利用基色和混合专色薄膜印品在黑白两色基底上的反射光谱, 建立具有高透低反特性薄膜印品透射光谱的获取方法, 并利用透射光谱求解吸收光谱; 然后对专色吸光度与基色浓度进行回归分析, 确定吸光度与基色浓度线性相关度高的特征波段, 再根据郎伯比尔定律, 利用特征波长处的专色样本吸光度与基色样本吸光度建立组分预测模型, 求解专色样本的配方比例; 最后, 分析预测配方与预设浓度的偏差, 并按照预测配方浓度重新打样, 采用色差与光谱均方根误差对计算配方色与目标专色的匹配程度进行分析, 验证模型精度。 以凹印PET薄膜双基色混合样品为实验对象, 对提出的方法进行测试。 光谱分析表明, 基色样本和专色样本光谱反射率曲线在黑白两种基底上有明显差异, 但都随基色油墨浓度变化具有相同的变化规律, 专色样本的光谱透射曲线随着基色浓度的变化接近浓度比例大的基色曲线, 在400～580 nm区间, 专色的吸收光谱随着基色油墨A浓度降低而升高, 在580～700 nm区间随着基色油墨A浓度降低而降低。 除570～590 nm吸收曲线相交区间以外, 在可见光范围内专色油墨吸光度与基色油墨浓度之间的判定系数R2均大于0.95, 平均值为0.990 0, 两者具有强线性相关关系。 分别选取基色A和B波长为520 nm(R2为0.994 2)和700 nm(R2为0.998 5)处的吸光度代入配色模型, 并利用最小二乘法求解目标专色的配方比例。 实验结果表明, 六组预测浓度与预设浓度相比较, 配方平均偏差为2.5%, 无显著偏差。 配方色样本与专色样本色差最大值为1.98, 最小值为0.30, 平均值为0.85, 均满足GMI对专色复制的要求, 其中5组专色色差小于1.5, 符合专色忠实复制的要求; 6组专色的光谱均方根误差最大值为2.93%, 最小值为0.49%, 平均值为1.40%, 基本实现了同色同谱的高精度颜色复制。 验证了该方法对于提高专色油墨配色精度和改善PET薄膜印刷质量具有显著效果, 可为PET薄膜专色配方预测提供科学方法。

A model for predicting spot color ink formula for PET film printing is proposed based on the correlation between absorbance and the concentrations of primary inks. Firstly, according to the methodology with the concept of light propagation and the multiple internal reflectance of ink layer and substrate, a method for obtaining the spectral transmittance of the thin film prints with high transmission characteristic is established by making use of the reflectance of the film prints with back and white substrates. The absorption spectrum of the film prints in the visible spectrum is obtainedby the spectral transmittance. Then, the regression equation between the absorbance of spot color inks and the concentrations of a primary color ink is established to determine characteristic wavelengths with strong linear correlation. After that, according to Lambert-Beer law, the prediction model is established by using the absorbance of the spot color samples and the absorbance of the primary color samples at the characteristic wavelengths, in order to obtain the proportions of primary color inks. Finally, the deviation between the predicted formulas and the actual concentrations is analyzed. In order to evaluate the accuracy of the prediction method, the remade samples are compared with target samples in terms of chromatic difference and the spectralroot mean square error. Some dual-componentspot color samples produced on PET filmby gravureare used as experimental subjects to verify the proposed method. The analysis of spectrum shows that the reflectance on the black substrate is significantly different from the reflectance on the black substrate, but both have the same trend with the variation of primary ink concentration. The transmittance curves of spot colorsamples are situated between the curves of primary color samples and movecloser to the primary color with higher concentration. The absorption spectrum of samples increase in the regionof 400～580 nm with the decrease of concentration of primary ink A, and decrease in the region of 580～700 nm as the concentration of A decreases. Except for the region of 570～590 nm, the linear correlation coefficients R2 between absorbance of spot color inks and the concentration of primary color A are higher than 0.95 and the average value is 0.990 0, which means a strong linear correlation in the visible spectrum range. The absorbancevalues at the wavelength 520 nm (R2 of 0.994 2) and 700 nm (R2 of 0.998 5) of the primaryinks A and B are selected to predict the formulas of spot colors by using the least squares method. The results show that the 6 groups of predicted concentrations are with 2.5% deviation from the actual concentrations of target samples, which means no significant difference. The maximum chromatic difference between the targetsamples and remade samples is estimated to be 1.98, the minimum to be 0.30, and the average value to be 0.85, which satisfies the requirement of spot color reproduction. 5 of the 6 groups are smaller than 1.5, which satisfies the requirement of faithful reproduction. The maximum RMSE is 2.95%, and the minimum is 0.49%, and the average value is 1.40%, which means a high precision color reproduction in the visible spectrum. It is confirmed that the proposed method could effectively improve the printing quality and the spot color matching precision, which may provide a scientific method for the predictive study of spot color ink formula of PET film printing.