论文部分内容阅读
研究了山梨酸钾在水溶液和橙汁中的荧光特性,结果表明在两种溶液中山梨酸钾的荧光特性虽然有很大的区别,但是它们的荧光特征峰都存在于λex/λem=375/490nm。从二维荧光光谱可以看出,橙汁中山梨酸钾的浓度和相对荧光强度关系错综复杂,两者不再满足线性关系。为了准确测定橙汁中山梨酸钾的浓度,提出了一种微粒群(PSO)算法优化的误差逆向传播(BP)神经网络的新方法。两组预测浓度的相对误差分别为1.83%和1.53%,预测结果表明该方法具有可行性。在浓度范围为0.1~2.0g·L-1内,PSO-BP神经网络能够完成橙汁中梨酸钾浓度的准确测定。
Fluorescence characteristics of potassium sorbate in aqueous solution and orange juice were studied. The results showed that although the fluorescence characteristics of potassium sorbate in both solutions were quite different, their fluorescence peaks all existed at λex / λem = 375 / 490nm . As can be seen from the two-dimensional fluorescence spectrum, the relationship between the concentration of potassium sorbate in orange juice and the relative fluorescence intensity is complicated, and the two no longer satisfy the linear relationship. In order to accurately determine the concentration of potassium sorbate in orange juice, a new method based on Particle Swarm Optimization (PSO) neural network with error back propagation (BP) was proposed. The relative errors between the two groups were 1.83% and 1.53%, respectively. The prediction results show that this method is feasible. In the concentration range of 0.1 ~ 2.0g · L-1, PSO-BP neural network can accurately determine the concentration of potassium citrate in orange juice.