论文部分内容阅读
将主成分分析用于优化径向基函数神经网络的输入变量,用于提高神经网络模型的预测能力。方法能有效地解决分子荧光光谱法测定尿液中诺氟沙星过程中尿液中内源性荧光物质的干扰。在优化条件下,径向基函数神经网络模型对尿液中诺氟沙星的平均预测误差为15.34%,神经网络结构为2∶3∶1。方法已用于测定尿液中的诺氟沙星。
Principal component analysis is used to optimize the input variables of radial basis function neural networks to improve the predictive ability of neural network models. The method can effectively solve the interference of endogenous fluorescent substance in urine during the determination of norfloxacin in urine by molecular fluorescence spectrometry. Under the optimized conditions, the radial basis function neural network model has an average prediction error of 15.34% for norfloxacin in urine and a neural network structure of 2: 3: 1. The method has been used to determine norfloxacin in urine.