提出一种基于粗糙集理论的模糊神经网络系统，首先运用粗糙集理论来发现大量样本数据中的概略化的规则，然后根据这些规则来设计神经网络的结构模型，并利用神经网络技术对模型进行训练。化纤工业中抽丝冷却侧吹风过程的模拟仿真实验，证明了该方法的有效性和可行性。

A fuzzy neural network based on rough sets is presented in this paper. First, a set of rough rules are found from the given training data by using rough sets theory, then the structure and model are designed according these rules, and then the model is trained by neural network technique. The experiments that simulate the control process of side-wind for chemical fiber are carried out. The results proved its efficiency and feasibility.