【摘要】 将改进小波神经网络与BP神经网络相结合,提出一种新的混级联神经网络结构,用于单扫描示波极谱信号的同时测定。通过对网络结构的优化和网络参数的调整,加快了训练速度,提高了预测的准确度。用该法对邻、间硝基氯苯混合样进行了预测,结果满意。对级联神经网络法与单一BP神经网络法的预测结果进行了比较,表明级联神经网络优于单一BP神经网络。更多还原

【Abstract】 By combining the improved wavelet neural network with BP neural network,a new structure based on mixed cascade neural network was established.This new neural network have been used in the determination of single-sweep oscillopolarographic signals.The training speed and the predication accuracy can be enhanced by optimizing the network structure and parameters.Concentrations predicated for m-Nitrochlorobenzene(m-NCB in short) and p-Nitrochlorobenzene(p-NCB in short) mixture are satisfactory.All the results suggested that the cascade neural network method was better than the single BP neural network method.更多还原