同时,应用神经网络的学习和记忆功能,对控制变量的隶属函数和控制规则进行优化,使控制方案更趋于合理。
At the same time, using the study and memory ability of neural networks, optimize subject function and control rules of control variables, which makes the control scheme more reasonable.
为使模糊屈服面明朗化,文中引入了与模糊集截集相应的模糊屈服函数族。并巧妙地将隶属函数和屈服函数联系起来。
Fuzzy yield functions corresponding to he cut-set of fuzzy sets are introduced for the obviousness of fuzzy yield surface, and the membership function is ingeniously related with yield function.
选取的隶属函数使神经网络权值有一定的知识表示意义,并通过模糊化层将输入特征量转化为模糊量。
The selected membership function made neural network weight values have definite knowledge meaning, and the input characteristic variables were translated into fuzzy variables by fuzzy layer.
应用推荐