论文部分内容阅读
结合模糊集合理论，将结构化神经网络方法用于包装件缓冲垫层非线性特性识别问题．对于两种典型的包装件缓冲垫层材料模型的模拟识别结果表明，据此方法可以较好地获得其非线性特性．模糊自适应技术的引入，提高了网络训练速度，减少了对于训练参数的人为干预，使得结构化神经网络方法更适于实际应用．
Combining with fuzzy set theory, structured neural network method is applied to the nonlinear characteristic identification of package cushion layer. The simulation results of the two typical cushion cushion material models show that the method can obtain the nonlinear characteristics well. The introduction of fuzzy adaptive technology improves the speed of network training and reduces the human intervention for training parameters, making the structured neural network method more suitable for practical application.