自学习模糊逻辑推理网络及模糊控制器的构成
Formation of A Learning Fuzzy Logic Reasoning Network and A Fuzzy Controller
-
摘要: 提出了一种自学习模糊逻辑推理网络和自学习模糊控制器的构成方法.这种方法是 把RCE(Restricted Coulomb Energy)模型进行扩展.使其能够进行模糊逻辑推理,并用于构 成基于RCE模型的自学习模糊控制器RLFC(RCE—based Learning Fuzzy controller).这种 方法有以下特点:a)学习速度高.追加学习容易;b)网络的信息处理工作单元的个数由自学习 决定,通用性好;c)不存在局部极小点问题.自学习模糊控制器RLFC可以直接把熟练者的操 作知识转换成模糊控制规则,自动构成模糊控制器.数值仿真实验表明其效果良好.Abstract: This paper presents an RLFC (RCE-based learning fuzzy controller) which is capable of extracting expert knowledge automatically. The RLFC is an extended RCE(Restricted coulomb energy) model, hence it needs few iterations in learning and is easy for additional learning when control objects are changed. Moreover, the RLFC is capable of dealing with fuzzy sets and produces fuzzy control rules in a self-organizing way from operational patterns of experts. The effectiveness of RLFC is shown by numerical simulation results.
-
Key words:
- Learning network /
- fuzzy control /
- RCE model
计量
- 文章访问数: 2470
- HTML全文浏览量: 183
- PDF下载量: 1065
- 被引次数: 0