论文部分内容阅读
利用近红外光谱技术在线检测水果内部品质的关键是获取精度高稳健性好的定量分析模型。研究开发了短波近红外光谱苹果品质在线检测系统,试验时苹果样本传输速度为5个/s,以漫反射方式采集,有效光谱范围为500~1100 nm。经光谱强度标准化校正后,有比较的采用遗传算法、连续投影算法和蚁群优化算法等提取特征变量,分别建立偏最小二乘模型,同时分析了这三种方法提取光谱特征变量的搜索机制。特征变量提取方法建立的预测模型所用变量显著减少,预测效果均优于全光谱模型,且能提高运算速度,增强模型的稳健性;其中又以蚁群优化算法的模型预测能力最佳,预测集相关系数R为0.9358,预测均方根误差RMSEP为0.2619。研究结果表明,近红外光谱结合特征变量提取方法可以建立高效的苹果可溶性固形物含量在线检测模型,在产业化应用方面具有很大潜力。
The key of using NIRS to detect the internal quality of fruit is to obtain the quantitative analysis model with high precision and good robustness. Research and development of short-wave near-infrared spectroscopy apple online detection system, the apple sample transmission speed of 5 / s, collected by diffuse reflection, the effective spectral range of 500 ~ 1100 nm. After the spectral intensity was normalized and corrected, some feature variables were extracted by using genetic algorithm, continuous projection algorithm and ant colony optimization algorithm to establish partial least squares model respectively. At the same time, the search mechanism for extracting spectral characteristic variables by these three methods was analyzed. The variables used in the feature variable extraction method significantly reduce the variables used in the prediction model, and the prediction results are better than the full-spectrum model, and the computational speed and the robustness of the model can be improved. In addition, the ant colony optimization algorithm has the best model prediction ability and the prediction set The correlation coefficient R is 0.9358 and the root mean square error of prediction RMSEP is 0.2619. The results showed that near-infrared spectroscopy combined with eigen-variable extraction method could establish an efficient on-line detection model of apple soluble solids content, which has great potential for industrial application.