中文摘要：

结论参数对T-S模糊模型的泛化能力有重要影响.该文引入v-支持向量回归机（v-SVRM）,把T-S模型结论参数的辨识问题转化为一个约束优化问题,并推导了新的迭代求解算法.该方法通过一个参数v控制支持向量的数目和落在？？不灵敏带外样本点的数目,并自动计算合适的？？.针对典型负荷被控对象的仿真结果表明：该方法比通常采用最小二乘法进行结论参数辨识的方法具有更好的泛化能力;此外,由于采用了？？不灵敏损失函数,该方法具有更好的噪声适应能力.

英文摘要：

This paper introduced v-support vector regression machine to identify consequent parameters which affect generalization performance in T-S fuzzy model. It was finally converted to an optimizing problem with inequalities constraint, and a new efficient iterative algorithm was presented to solve it. In this method, a parameter v lets one effectively control the number of support vectors and the number of points that come to lie outside of the so-called e-insensitive tube, moreover, a suitable e can be determined automatically. Simulations on a typical load system indicate that this method has improved generalization ability than methods that use least square algorithm. In addition, it is more tolerant to noise data for e-insensitive loss function is used.