一种T-S模糊系统建模方法

A T-S Fuzzy System-based Modeling Method

T-S模糊系统被广泛应用于基于数据的建模应用中。模糊规则作为系统的核心,是影响系统性能的重要因素。在分析常见模糊系统建模方法的基础上,提出一种简单有效的建模方法。该算法基于变结构模糊建模思想,均匀选择模型的初始结构,以绝对误差为建模指标,通过增加模糊规则来提高T-S模糊系统的精度。为降低规则参数辨识的计算量,提高建模速度,将规则参数分为线性和非线性两部分,分别采用不同方法进行辨识。实例证明文中所提出的建模方法规则分布合理,收敛速度快,建模精度高,具有很好的实际应用价值。

As a method of data-driven modeling, T-S fuzzy system has been used in a wide variety of applications. However, inappropriate fuzzy rule sets can decrease the precision and generalization of the system. A new modeling method was proposed to establish T-S fuzzy model. The new method was based on the idea of variable structure, i.e., initial rules were designed uniformly, and new fuzzy rules were added to reduce the maximum absolute error （MAE） index. In order to accelerate the computational convergence of the parameter optimization, the model parameter optimization depends partly on the Levenberg-Marquardt method for nonlinear parameter optimization and partly on the least squares method for linear parameter estimation. This method makes the rule distribution rational, computational convergence fast, and high model precision. The effectiveness of the proposed method is illustrated by a number of simulation examples.