论文部分内容阅读
基于罗丹明6G的分子荧光原理,通过对比不同实验条件下得到的罗丹明6G荧光光谱,得出p H为1条件下的相对荧光强度最大。罗丹明6G试剂中加入钼酸铵、磷酸二氢钾、硫酸试剂生成络合物后,罗丹明6G的相对荧光强度值有所下降,在一定范围内表现出线性关系,罗丹明6G荧光峰的位置没有发生变化。基于遗传算法-逆向误差传播(GA-BP)神经网络构建了输入节点数为36×18的矩阵、输出节点数为1×18的矩阵、以检测磷酸盐浓度为目的的非线性模型。网络训练中,误差精度为10-3,输出与期望的相关系数为0.998,网络预测中,平均回收率为99%,平均标准偏差值为1.79%,达到了理想的检测效果。证明此网络适用于检测0~2.00 mg/L的磷酸盐溶液。提供了一种快速、有效检测磷酸盐浓度的方法,有助于环境检测技术的发展和应用。
Based on the molecular fluorescence principle of rhodamine 6G, by comparing the fluorescence spectra of rhodamine 6G obtained under different experimental conditions, it is found that the relative fluorescence intensity under the condition of p H is 1 is the maximum. Rhodamine 6G reagent by adding ammonium molybdate, potassium dihydrogen phosphate, sulfuric acid reagent to generate complex, the relative fluorescence intensity of rhodamine 6G decreased, showing a linear relationship within a certain range, rhodamine 6G fluorescence peak No change in location. A non-linear model for the purpose of detecting phosphate concentration was constructed based on GA-BP neural network with a matrix of 36 × 18 input nodes and a matrix of 1 × 18 nodes. In network training, the accuracy of error was 10-3, and the correlation coefficient between output and expectation was 0.998. In network prediction, the average recovery was 99% and the average standard deviation was 1.79%, which achieved the ideal detection result. Prove that this network is suitable for the detection of 0 to 2.00 mg / L phosphate solution. It provides a quick and effective method to detect phosphate concentration and contributes to the development and application of environmental detection technology.