论文部分内容阅读
依据小波函数的非线性逼近能力和神经网络的自学习能力,用Morlet小波基函数代替传统的神经网络传递函数Sigmoid函数,克服BP神经网络易陷入局部极小等缺陷,建立了小波神经网络算法,用于紫外分光光度法不经分离同时测定复方苯甲酸酊中的苯甲酸和水杨酸。苯甲酸和水杨酸的平均加标回收率和RSD分别为101.1%,1.7%;102.1%,1.8%。结果表明,该方法不需要物理化学分离,分析速度快,精确度较高,适用于该制剂的含量测定。
According to the non-linear approximation ability of wavelet function and the self-learning ability of neural network, Morlet wavelet basis function is used to replace the traditional Sigmoid function of neural network transfer function to overcome the defects that BP neural network is easy to fall into local minima. A wavelet neural network algorithm is established, For the simultaneous determination of benzoic acid and salicylic acid in compound benzoic acid tincture without separation by ultraviolet spectrophotometry. The average spiked recoveries and RSDs of benzoic acid and salicylic acid were 101.1%, 1.7%, 102.1% and 1.8%, respectively. The results show that this method does not require physical and chemical separation, analysis speed, high accuracy, suitable for the determination of the content of the preparation.