改进型B样条网络函数及其应用

B样条网络函数可以作为模糊神经系统的隶属函数，通过指出常规B样条模糊隶属函数构造中存在的不足，提出了针对性的改进措施，从而既满足了B样条本身的性质，又从理论设计上避免了使系统不稳定的因素，增加B样条函数在作为模糊神经网络控制器的实际应用中的抗干扰能力，论给出了改进型B样条网络的设计过程以及实验结果，在实际电机速度控制系统的对比实验中证明了所提改进方法的有效性和实用性。

An Improved B-spline Network Function and Its Application

B-splinefunction can be used as the membership func tion in the fuzzy neural system. This paperproposes an effective way in order to overcome the shortages in the construction ofB-spline fuzzy membership func tions, which both satisfies the properties of B-splineitself and avoids factor s of system instability from the theory design point, thusincreasing the abilit y of the resisting disturbances of the B-spline function used as afuzzy neural network control in actual applications. The paper also gives the designprocedu re and the experiment results of the improved B-spline networks. The effectiveness and practicality of the proposed strategy have been verified through an actu al motorvelocity control system experiment.