Abstract: According to shortcomings of traditional fuzzy controller prone to rule explosion in the control process, this paper puts forward a method which is based on vector “membership” fuzzy controller struc-ture optimization. This optimization method is based on the improvement of the traditional fuzzy controller’s membership and degree level. The current inputs relative direction and size gradation reflect in vector “membership” and scalar fuzzy level, thus can greatly reduce the number of fuzzy rules, to reduce the structure complexity and improve the control efficiency. The simulation of DC-motor control effect shows that this kind of controller has a simple structure and smaller error than traditional controller, and also proves the feasibility and effectiveness of this method.

Key words: rule explosion, structure optimization, vector “membership”

摘要： 针对传统模糊控制器在控制过程中容易发生规则爆炸的缺点，提出一种基于矢量“隶属度”的模糊控制器结构优化方法。这种优化方法通过对传统模糊控制器的隶属度和模糊等级进行改进，把当前输入量的相对方向和大小等级分别反映在矢量“隶属度”和标量模糊等级上，从而能够大大减少模糊规则数目，降低结构复杂度，提高控制效率。直流电动机的仿真控制效果表明，矢量“隶属度”法优化后的控制器比传统控制器结构简单和误差小，从而验证了这种方法的有效性和可行性。

关键词: 规则爆炸, 结构优化, 矢量&ldquo, 隶属度&rdquo