论文部分内容阅读
传统的RBF神经网络在逼近能力、分类能力和学习速度等方面均优于BP网络,但为了使RBF神经网络的收敛速度和网络精度等更好地满足实际需求,用到一个线性非线性并列新型结构的RBF神经网络模型,并将该模型应用到纺织物品的染色配色问题上。应用该模型对染料的浓度与CMY值进行配色计算,实验表明具有较好的效果。改进后的RBF神经网络所表现出的良好性能,为其在该领域的应用提供了参考。
The traditional RBF neural network is superior to BP network in the aspects of approach ability, classification ability and learning speed, but in order to make the RBF neural network convergence speed and network accuracy better meet the actual needs, a new linear nonlinear parallel Structure of the RBF neural network model, and the model is applied to the textile dyeing and color issues. Applying the model to calculate the dye concentration and CMY value, the experiment shows that it has a good effect. The good performance of the improved RBF neural network provides a reference for its application in this field.